Characterization of Iron-Responsive Transcription Regulators In

CHARACTERIZATION OF IRON-RESPONSIVE TRANSCRIPTION REGULATORS IN

PYROCOCCUS FURIOSUS

by YIXUAN ZHU Under the Direction of ROBERT A. SCOTT and MICHAEL W.W. ADAMS

ABSTRACT

Iron is an essential element for the growth of Pyrococcus furiosus since limited medium iron causes impaired growth. In previous iron-limitation microarray experiments, two putative iron transporters Ftr1 and FeoAB were significantly up-regulated in response to limited iron in the growth medium. This change in the iron acquisition system raises the question: what is responsible for the iron-dependent regulation? Two putative iron-responsive transcription factors were found in P. furiosus, Fur and DtxR. Studies on Fur-DNA binding activity and the characterization of its deletion mutant strain revealed no demonstrable role for the Fur homolog in the regulatory response of the cells to iron. Analysis of the DtxR genomic sequence suggested an incorrect translation start site prediction by NCBI and TIGR; another start site resulted in a 12 amino acid extension on the N-terminus. Recombinant DtxR proteins were expressed (with "full- length" and without the N-terminal extension) and their DNA-binding affinity was determined using electrophoretic mobility shift assay. The full-length DtxR was verified to bind specifically to ftr1 and feoAB promoters, indicating the function of DtxR as a potential transcriptional factor.

The truncated DtxR failed to recognize any of the promoters. The dtxR deletion mutant (DTXR) did not show any growth phenotype compared to the control strain COM1C2 under iron-rich and iron-limited conditions. Microarray analysis of the iron-dependent regulation in DTXR and

COM1C2 provided evidence that DtxR controls the expression of Ftr1 and FeoAB in P. furiosus since both iron transporters were no longer iron-regulated in the deletion strain. Results from the quantitative PCR experiment confirmed that DtxR is an iron-dependent repressor for the putative iron transporters in P. furiosus. Meanwhile, the ftr1 deletion mutant (FTR1) was constructed to investigate its role in iron uptake. However, no phenotype was observed for its growth in iron- limited and metal-limited medium; the similar intracellular iron content detected by ICP-MS in

FTR1 and parent strain indicates that Ftr1 might not be an essential iron transporter despite its highly iron-responsive expression in P. furiosus.

INDEX WORDS: Archaea, Pyrococcus furiosus, transcription factor, transcriptional

regulation, iron acquisition, Fur, DtxR, DNA microarray, EMSA, deletion

mutant, ICP-MS.

CHARACTERIZATION OF IRON-RESPONSIVE TRANSCRIPTION REGULATORS IN

PYROCOCCUS FURIOSUS

by YIXUAN ZHU B.S., Tsinghua University, 2005

A Dissertation Submitted to the Graduate Faculty of The University of Georgia in Partial

Fulfillment of the Requirements for the Degree

DOCTOR OF PHILOSOPHY

ATHENS, GEORGIA

2011

YIXUAN ZHU

CHARACTERIZATION OF IRON-RESPONSIVE TRANSCRIPTION REGULATORS IN

PYROCOCCUS FURIOSUS

YIXUAN ZHU

Major Professor: Robert A. Scott Michael W.W. Adams

Committee: I. Jonathan Amster John Rose

Electronic Version Approved:

Maureen Grasso Dean of the Graduate School The University of Georgia December, 2011

DEDICATION

To my parents who gave life to me, love me and support me for all time.

To my dear grandfather who has been an inspiration to me, passed away during my pursuit of Ph.D., R.I.P.

ACKNOWLEDGEMENTS

I would like to thank my lab mate Gina Lipscomb, who shared four years time with me in three different labs, for always being patient and helpful and offering invaluable advice on my research; Gerrit Schut and Angeli Menon, for providing important experimental data and sharing useful research experiences; Karen Stirrett, for constructing the strains used in this research;

Alex Cvetkovic and Sunil Kumar for running ICP-MS for my samples and providing insightful suggestions; Chris Hopkins and Junsong Sun for teaching me how to use the ÄKTA purifier and giving me advice on protein purification; Farris Poole who offered me firsthand training when I joined the lab and technical support along with my research; and Darin Cowart for providing the database to search the potential DtxR binding site.

I am also thankful to Dr. Eidsness from whom I got much support and encouragement through my first year in graduate school; Dr. Westpheling and Dr. Adams who offered me workspace in their labs and wonderful suggestions on my research; and my advisor Dr. Scott, not only for the guidance to my works, but also for the patience and support that helped me overcome crisis situations and finish my dissertation.

And last but not least, I express my gratitude to all those that I have been honored to work with, including previous and current lab members, faculty and departmental staffs from the departments of Chemistry and Biochemistry.

TABLE OF CONTENTS

Page ACKNOWLEDGEMENTS ...... v

CHAPTER

1 OVERVIEW OF TRANSCRIPTIONAL REGULATION IN ARCHAEA ...... 1

1.1 Introduction to archaea ...... 1

1.2 Archaeal basal transcription...... 2

1.3 Archaeal transcriptional regulation and regulatory transcription factors ...... 5

1.4 The model archaeon Pyrococcus furiosus and its iron homeostasis ...... 10

2 MATERIALS AND METHODS ...... 15

2.1 Protein expression and purification ...... 15

2.2 Gel filtration to determine protein quaternary structure ...... 16

2.3 Electrophoretic mobility shift assay...... 17

2.4 Strains and growth conditions ...... 18

2.5 Mutant strain construction ...... 19

2.6 Growth characterization of cultures ...... 21

2.7 Quantitative PCR ...... 22

2.8 Microarray transcriptomics and data analysis ...... 22

2.9 Determination of intracellular iron content...... 24

3 CHARACTERIZATION OF THE PUTATIVE IRON-RESPONSIVE TRANSCRIPTION FACTOR FUR ...... 25

3.1 Introduction to the Fur family protein ...... 25

3.2 Fur homolog in Pyrococcus furiosus ...... 29

vii

3.3 Validation of the DNA-binding affinity of PfFur ...... 30

3.4 Characterization of the fur deletion mutant ...... 32

4 CHARACTERIZATION OF THE PUTATIVE IRON-RESPONSIVE TRANSCRIPTION FACTOR DTXR ...... 58

4.1 Introduction to the DtxR family protein ...... 58

4.2 DtxR homolog in Pyrococcus furiosus ...... 60

4.3 Reanalysis of the DtxR homolog in Pyrococcus furiosus ...... 63

4.4 Validation of the DNA-binding affinity of PfDtxR ...... 64

4.5 Characterization of the dtxR deletion mutant...... 70

5 CHARACTERIZATION OF THE PUTATIVE IRON PERMEASE FTR1 DELETION MUTANT ...... 120

5.1 Introduction ...... 120

5.2 Characterization of the ftr1 deletion mutant ...... 121

6 CONCLUSIONS...... 134

6.1 Effect of iron on the growth of Pyrococcus furiosus ...... 134

6.2 PfDtxR-DNA binding affinity ...... 135

6.3 Physiology of DTXR mutant and the mode of regulation by DtxR ...... 136

REFERENCES ...... 138

APPENDIX ...... 162

vii

CHAPTER 1

INTRODUCTION

1.1 Introduction to archaea

The Archaea are microorganisms evolved as one of the three primary lineages several billion years ago. The first scientific literature came out describing the discovery of archaea dated about 200 years ago, and the Archaea were not formally proposed as the third domain of life till the 1970s [1, 2]. Molecular, genomic and phylogenetic evidences have been accumulated ever since then to strengthen the definition of Archaea as a third domain of life in addition to

Bacteria and Eukarya [1, 3-6]. One unique feature found for many members of this kingdom is their ability to thrive in physically or geochemically extreme conditions with a great diversity of temperature, pH, salinity and pressure, ranging from harsh environments, such as hot springs and salt lakes to soils, oceans, marshlands and the human colon [7, 8].

While most of the metabolic and morphologic aspects of Archaea are similar to those of

Bacteria; the transcription and translation processes in Archaea appear to be a mosaic of bacteria and eukaryotes, containing factors closely related to those previously characterized from both

Eukarya and Bacteria. For the transcription apparatus, the fundamental design of the archaeal basal transcription machinery can be considered as a similar but simplified version of that of

Eukarya [9-11]; however, the components involved in the regulation of transcription and the mechanisms understood in Archaea are by far more frequently shared with that in Bacteria [12].

The relative simplicity of the archaeal transcription system, together with the inherent stability of many archaeal proteins, makes this a very attractive system to study the fundamental

1 mechanisms of transcription and regulation processes. Studies on the regulation of archaeal transcription should provide a wealth of data which will be helpful for getting a better understanding of the regulatory mechanisms in all three domains of life.

1.2 Archaeal basal transcription

Transcription is the process that involves the transcribing of genetic information from

DNA to RNA. The transcribed DNA message is used to produce proteins. DNA is transcribed by

RNA polymerase, which attaches to a specific area called promoter region with the assistance of transcription factors, and there are specific nucleotide sequences that tell RNA polymerase where to begin and where to end transcription. The last decade marked a revolution in understanding the molecular detail of cellular RNA polymerases (RNAPs) from bacteria, archaea, and eukaryotes and their respective transcription mechanisms. The archaeal transcription system has been characterized as a hybrid of eukaryotic and bacterial transcription systems [13]; as the archaeal basal transcription apparatus is very similar to that of eukaryotes [14, 15], its transcriptional regulatory factors are more similar to those of bacteria [16, 17]. The archaeal transcription machinery is closely related to the eukaryotic systems in terms of RNAP structure and function, but relies on a minimal set of transcription factors and promoter elements. Similar to the structures of Saccharomyces cerevisiae RNAP II which include 10-12 subunits [18], archaeal RNAPs, depending on species, generally consist of 11-13 subunits (A‟+A‟‟, B, D, K, L,

F, H, E, G, N, P and Rpo13) [14, 19] that are structurally and functionally homologous to the subunits of eukaryotic RNAPII. On the other hand, bacterial RNAP only contains five subunits:

αI, αII, β', β'', and ω, which are all essential for RNAP cellular function [20]. The X-ray crystal structures of archaeal RNAPs have been determined from two Sulfolobus species [21-23] and the cryoelectron microscopy structure has been determined from Pyrococcus furiosus [24]. The

2 overall architecture of bacterial, archaeal and eukaryotic RNAP resembles a “crab claw”, but there are significant differences on the surfaces of these structures. The most significant difference is that archaeal RNAP and all three types of eukaryotic RNAPs have a protruding stalk-like structure that is absent from bacterial RNAP [18, 21, 25, 26].

In order to form a transcription-ready open complex from the preinitiation complex (PIC),

RNAP unwinds double-stranded DNA extending from ~11 base pairs upstream to a few base pairs downstream from the transcription start site. This process may require additional protein factor(s), which can vary depending on the organism. In bacteria, the σ factor binds to the core

RNAP to form the holoenzyme, which recognizes promoter DNA at around -35 and -10 from the transcription start site and makes a closed RNAP-promoter complex essential for DNA unwinding [27]. In RNAP II-dependent transcription, six general transcription factors, TFIIA,

TFIIB, TFIID, TFIIE, TFIIF, and TFIIH, mediator complex and RNAP II are recruited to the promoter DNA and form the PIC; and TFIIF and TFIIH are known to play several critical roles in open complex formation [28]. In archaea, the catalytic core of archaeal RNAP A‟+A‟‟ subunits does not contain the C-terminal domain (CTD) common to its eukaryotic counterpart and therefore lacks a target for many regulatory factors. As a result, the archaeal PIC is similar to the eukaryotic RNAP II transcription system but contains only TBP (TATA-binding protein),

TFB (Transcription Factor B), and TFE (Transcription Factor E), which are orthologs of TBP,

TFIIB, and TFIIEα, respectively. TBP and TFB are the only proteins required for PIC formation in archaea; these proteins play a role in recruiting RNAP to the promoter. Many archaea encode multiple homologs of TBP and TFB. Halobacterium NRC-1 encodes six TBP and seven TFB variants that allow multiple combinations of different TFB-TBP which could facilitate differential regulation of transcription by alternative TFB-TBP combinations [29]. Thermococcus

3 kodakarensis encodes two TFB variants, TFB1 and TFB2, and combined genetic and biochemical approaches have shown that both are active and support transcription from a range of promoters [30].

TBP binds promoter DNA along the minor groove of an AT-rich TATA-box sequence located ~25 bp upstream of the transcription start site and causes a sharp DNA bend [31-33]. The distortion of the DNA backbone is thought to assist the recruitment of other transcription factors to the promoter. Different consensus sequences have been established for the TATA-box in different branches of the Archaea; TTTTTAAA in Crenarchaea (predominantly in Sulfolobus species), TTTATATA in methanogens, TTTWWW in halophiles, and TTWWWAW (W = A or

T) in Pyrococcus.

TFB interacts with a DNA sequence known as the B-factor recognition element (BRE), which is a purine rich (within nontemplate strand) sequence found immediately upstream of the

TATA-box. The interaction between TFB and BRE appears to be necessary to determine the direction of transcription [34, 35]. Previous in vitro transcription analyses showed that once 9-12 bases of the nascent transcript are formed, TFB is released from the transcription initiation complex and recycled for the next round of transcription [36, 37]. TFB release from RNAP is also necessary to open the channel for nascent RNA exit from RNAP [38].

The third archaeal initiation factor, TFE, associates with RNAP and stimulates DNA melting and loading. It does not contribute directly to the mechanism of transcription initiation, but interacts with the non-template strand and thereby stabilizes the PIC and remains associated with RNAP during elongation [39].

The ortholog of C-terminal domain of the eukaryotic elongation factor TFIIS was found in archaea as TFS. TFS plays a role in transcription proofreading by RNA hydrolysis [40]. No

4 orthologs to the eukaryotic TFIIA, TFIIF, TFIIH or TAF (TBP-associated factors) have been identified in archaea. In T. kodakarensis KOD1, a TBP-interacting protein (TIP26) has been identified to bind TBP and inhibit it from binding DNA [41, 42], indicating the possibility that other proteins with analogous functions might exist in other archaeal species.

1.3 Archaeal transcriptional regulation and regulatory transcription factors

In order to effectively survive in a competitive environment, organisms need to balance production of gene required for cell activities, while avoiding extraneous production of unnecessary proteins. The regulation of gene expression could potentially take place at any stage from transcription initiation to protein degradation. Regulation of transcription initiation is one of the important mechanisms governing gene expression.

Although archaea possess a basal transcription machinery resembling that of eukaryotes, their regulators are more homologous to bacterial activators and repressors [43]. To date, only a few archaeal regulatory systems have been characterized at the molecular level. Among the regulators that have been characterized in detail, most of them have been demonstrated to be transcriptional repressors [44].

MDR1 is a homolog of the bacterial DtxR family protein and is one of the first characterized transcription factors in Archaea. It is a metal-dependent regulator of

Archaeoglobus fulgidus [45]. MDR1 binds to its recognition sequence which overlaps the transcription initiation site and represses transcription from its own gene in the presence of metal ions by blocking recruitment of RNAP.

LrpA, a member of the bacterial Lrp/AsnC family from P. furiosus, negatively regulates the transcription of its own gene via binding to a 46-bp sequence that overlaps the transcription start site preventing recruitment of RNAP [46]. Several other Lrp family putative regulators have

5 been characterized: SaLrp from Sulfolobus acidocaldarius [47], SsLrp from S. solfataricus [48],

SsLrpB from S. solfataricus [49, 50], and Lrs14 from S. solfataricus [51]. Lrs14 represses transcription from its own gene by obstructing the binding of TBP and TFB through overlapping the BRE and TATA box.

The members of the Lrp family are also known as feast/famine regulatory proteins that have the potential to form higher-order nucleoprotein structures with DNA [52]. Bacterial regulators of the Lrp family have been shown to respond to small-molecule ligands, such as amino acids. An archaeal Lrp-like regulator, LysM from S. solfataricus, was found to respond to lysine. In the absence of lysine, LysM binds to a site upstream of the BRE/TATA box in the promoter of a lysine-biosynthesis gene cluster and most likely activates transcription; in the presence of lysine, LysM has lower DNA-binding affinity, and the expression of the gene cluster was reduced [53].

Another characterized archaeal repressor was TrmB, which does not appear to have any homologs either in bacteria or eukarya. TrmB is the transcriptional repressor for the gene cluster of the trehalose/maltose ABC transporter of the hyperthermophilic archaea Thermococcus litoralis and P. furiosus (the gene is identical in both organisms), with maltose and trehalose acting as inducers [54]. TrmB of Pyrococcus furiosus was discovered as the trehalose/maltose- specific repressor for the genes coding the trehalose/maltose high-affinity ABC transporter [55].

The determined structure of the C-terminal sugar-binding domain reveals a novel sugar-binding fold of TrmB [56].

Two related archaeal transcriptional regulators have also been described since the discovery of TrmB, TrmBL1 from P. furiosus [57] and Tgr from Thermococcus kodakarensis

[58]. TrmBL1 appears to be a global transcriptional regulator. It preferentially recognizes the

Thermococcales-glycolytic motif sequence that is found upstream of the high-affinity maltodextrin-specific ABC transporter as well as genes encoding enzymes involved in the glycolytic and the gluconeogenic pathway. It responds to maltose and maltotriose as inducers and functions as a repressor [55]. Tgr is an ortholog of TrmBL1, recognizing the same DNA- binding motif, functioning as both an activator and repressor of transcription in the hyperthermophilic archaeon Thermococcus kodakarensis. The tgr deletion mutant of T. kodakarensis showed growth defect under gluconeogenic conditions compared with the wild type strain. Microarray analysis of the tgr deletion mutant revealed the derepression of almost all genes related to glycolysis and maltodextrin metabolism [58]. The TrmBL2 protein in T. kodakarensis acts as a general chromosomal protein as well as global transcription repressor. It binds both coding and intergenic regions and represses transcription when bound to the promoter region [59].

The nitrogen repressor, NrpR, from Methanococcus maripaludis, is another archaeal- specific transcriptional regulator. NrpR controls the expression of the nitrogen fixation (nif) operon, resulting in full repression with ammonia, intermediate repression with alanine, and derepression with dinitrogen. NrpR binds to two tandem operators in the nif promoter region, nifOR1 and nifOR2 [60, 61].

The GvpD and GvpE transcriptional factors, which regulate the transcription of genes involved in gas vesicle formation, have been investigated in the halophilic archaeon

Halobacterium salinarum and Haloferax mediterranei. The transcriptional activator protein

GvpE resembles a basic leucine-zipper (bZIP) protein, and mutations in the putative DNA binding domain and in conserved residues of the leucine-zipper helix in GvpE, result in the complete loss of its activator function [62]. GvpD appears to be involved in the repression of gas

7 vesicle formation as a secondary regulatory protein in Haloferax mediterranei [63]. The GvpD and GvpE proteins of H. mediterranei are able to interact both in vitro and in vivo to regulate gas vesicle formation [64].

Phr, a repressor for heat shock genes in Pyrococcus furiosus, is the first characterized heat shock transcription factor in archaea. Phr represses the transcription of genes by abrogating

RNA polymerase recruitment to the TBP/TFB complex without affecting the binding of

TBP/TFB to the promoter. DNaseI footprinting analyses showed that Phr recognizes a palindromic nucleotide sequence: 5‟-TTTN2TN2CN5G N2A N2AAA-3‟ (N=A, T, G or C), which is conserved in heat shock promoters in P. furiosus and Pyrococcus abyssi [65]. Phr DNA- binding domain shows a tertiary fold similar to those in proteins belonging to the bacterial

SmtB/ArsR family, indicating a common molecular ancestor of archaeal and bacterial transcriptional factors [66].

Ptr2 is a transcriptional activator characterized in Methanocaldococcus jannaschii and

Methanocaldococcus thermolithotrophicus which regulates the expression of ferredoxin, rubrerythrin and proteins involved in electron transport. It is a homolog of the Lrp/AsnC family of bacterial transcription regulators and has been shown to activate transcription by its conjugate core transcription apparatus in vitro [67]. Activation of transcription is exerted through the direct interaction of Ptr2 with TBP, thus leading to a stimulated recruitment and probably also by affecting post-recruitment steps. Mutating residues in TBP have been shown to negatively affect activation by Ptr2 [68].

Another activator, Sta1, was isolated from Sulfolobus islandicus. Sta1 was able to bind promoters independently of any component of the PIC, its activating effect on transcription initiation was demonstrated in in vitro transcription experiments [69].

The sulfur (S0) -responsive regulator SurR was identified and characterized in

Pyrococcus furiosus [70]. It is an activator for its own gene and two hydrogenase operons whose expression is down-regulated during the primary S0 response. It is also a repressor for the primary S0-reducing enzyme NAD(P)H sulfur reductase (NSR) which is up-regulated during the primary S0 response. The activity of SurR is modulated by cysteine residues in a CxxC motif that constitutes a redox switch. Oxidation of the motif mediated by S0 inhibits specific DNA-binding by SurR [71].

There is now an increasing body of data available on the DNA binding properties of several putative transcriptional regulators in the archaeal domain of life. Repressors and activators prevent or facilitate recruitment of basal machinery to the core elements, respectively.

However, little is known about the cofactor requirements, in vivo mechanisms, and targets of many of these regulators. The answers to these questions require advanced investigation and further exploration.

1.4 The model archaeon Pyrococcus furiosus and its iron homeostasis

Pyrococcus furiosus has emerged as a useful model archaeon for focused study. It is one of the hyperthermophiles which comprise a group of organisms that grow optimally at or above

80 °C, including 34 genera from archaeal and bacterial domains [72]. P. furiosus is a member of the family Thermococcaceae. It was discovered by Fiala and Stetter from a hydrothermal vent community off the coast of Italy. The optimal growth temperature for P. furiosus is 100 °C; both carbohydrates and peptides can be used as carbon sources by P. furiosus [73]. The medium used in our laboratory to grow P. furiosus contains 44 metals – seven added metals and the remaining coming from added organic components. A study of the stable cytoplasmic metalloproteins in P. furiosus using liquid chromatography, high-throughput tandem mass spectrometry (HT-MS/MS)

9 and inductively coupled plasma mass spectrometry (ICP-MS) found that P. furiosus only specifically assimilated 21 of the 44 metals, with Fe being the most abundant metal of all [74]. A number of Fe-containing proteins involved in H2 production, energy metabolism, electron transfer and anaerobic detoxification pathways have been purified and characterized from P. furiosus [75-91]. The existence of these proteins suggests a crucial role of Fe in P. furiosus cellular metabolism.

Furthermore, two ferritin-like genes are found in the P. furiosus genome. The ferritin homolog (PF0742) is structurally similar to known bacterial and eukaryotic ferritins; it is a 24- mer of 20 KDa subunits, could contain 2,700 Fe when fully loaded, and is extremely thermostable. The wild-type ferritin is the only Fe-containing protein observed in a native protein gel of P. furiosus cell extracts stained with Prussian blue for iron, indicating its potential role as an iron-storage protein [77, 92]. The other ferritin-like protein is Dps, a member of a protein family that functionally manage the toxicity of oxidative stress by sequestering intracellular ferrous iron and using it to reduce hydrogen peroxide in a two electron process to form water.

Twelve copies of PfDps protein subunit self-assemble into a roughly 10 nm spherical cage-like quaternary structure; iron is converted to a benign form as Fe(III) within the protein cage. The transcription of dps is up-regulated in response to oxidative stress; the Dps-mediated reduction of hydrogen peroxide, coupled with the protein‟s capacity to sequester iron, may contribute to its service as a multifunctional antioxidant in P. furiosus [93-95].

Transcription analysis of cells grown with and without S0 revealed that PF2025, designated as the gene for S0-induced protein A (SipA), which encodes a 19 KDa protein, is significantly up-regulated as part of the secondary response [96]. It has been proven that the regulation of sipA is specifically due to sulfide rather than elemental sulfur and the regulation is

10 absolutely dependent on iron [97]. SipA is also found to be up-regulated in oxidative stress conditions [94, 95]; it assembles into a protein cluster and sequesters intracellular iron sulfide

(Sonya Clarkson, unpublished). SipA is therefore proposed to be the sulfide equivalent of Dps, playing a role in intracellular iron sulfide detoxification [97].

Since iron is an essential element for the growh of P. furiosus and a large number of genes in the P. furiosus genome are predicted to encode iron-containing proteins, it is reasonable to assume that P. furiosus has an efficient iron acquisition system to meet the organism‟s need for iron. A genomic analysis based on NCBI and TIGR databases showed a number of genes annotated as iron transporters in P. furiosus (Table 1.1). These include the homologs of high- affinity ferrous transporters FeoAB and Ftr1. The bacterial FeoAB homolog, is encoded by two genes in an operon, has been reported to play a role in metal homeostasis in several bacterial species [98-100]. Ftr1, the homolog of the eukaryotic high-affinity iron uptake permease FTR1, which couples with Fet3 oxidase in Saccharomyces cerevisiae to transport ferrous iron is also found [101], even though there is no ORF encoding the Fet3p in P. furiosus genome. A BLAST search revealed over 150 strains of bacteria with members of FTR1 superfamily. In E. coli, EfeU

(or YcdN) has been shown to be important for iron transport; it is part of a tripartite operon

(efeUOB) induced by low-pH and repressed by Fe2+-Fur, while the precise roles of the other two genes are unclear [102-104]. Although genomic analysis failed to find any siderophore synthesis systems, several ABC-type transporters were found in the genome suggesting P. furiosus may use siderophores secreted by other organisms for iron uptake. Comparison of the potential iron transport genes to those in the genetically closely related thermophile species T. kodakarensis shows a more compact version of the iron acquisition system in P. furiosus since six gene clusters in T. kodakarensis are annotated as putative iron transporters, but only three homologous

11 gene clusters (PF0502-PF0503, PF0909-PF0911 and PF1519-PF1520) could be found in P. furiosus.

According to previous iron-limitation microarray expression analysis (Angeli Menon, unpublished), PF0723 (ftr1) and PF0857 (feoB) were the only two genes annotated as iron transporters that were significantly up-regulated in response to iron limitation, by ~7 and ~4-fold, respectively, indicating that FTR1 and FeoAB may play an important role in P. furiosus iron acquisition. Given this observation, it is of interest to know the transcription factor(s) responsible for the regulation of the iron transporters in P. furiosus.

Several iron-responsive transcription regulators have been characterized in Bacteria and

Eukarya. This includes the widespread Fur family and DtxR family proteins. Fur family proteins have been characterized in many bacterial species and are the most well studied iron-responsive regulators to date [105]. Fur family proteins have a broad range of metal selectivity, members of the Fur family have been shown to specifically recognize Fe2+, Zn2+, Ni2+, Mn2+ and regulate the metabolism of the corresponding metal [106]. DtxR family proteins are mostly conserved in gram-positive bacteria with high GC-content and members of this family have been found to respond specifically to either Fe2+ or Mn2+ [107]. Rrf2 family proteins are poorly conserved in bacteria and use a variety of cofactors, including FeS [108]. The iron-responsive Aft1 is a transcription activator in Sacchromyces cerevisiae which functions to activate transcription of target genes in response to iron deprivation [109]. IRP (iron regulatory protein) is a post- transcription regulator in human and functions as an iron-responsive element-binding protein involved in the control of iron metabolism by binding mRNA to repress translation or degradation [110].

A BLAST search based on the protein sequences of the characterized iron-responsive transcription regulators in P. furiosus results in only homologs of Fur and DtxR family proteins being found within the Thermococcacea. We have therefore chosen to evaluate the role of Fur and DtxR in iron metabolism of P. furiosus.

Table 1.1 Pyrococcus furiosus ORFs potentially involved in iron transport

ORFa Name Annotationb Tk homologc Ferrous Transporter PF0723* ftr1 Iron permease FTR1 family -

PF0857* feoB ferrous iron transport protein b TK0714/TK0957 PF0858 feoA hypothetical protein TK0715/TK0958

ABC-type Iron Transporter PF0502 iron(III) dicitrate transport system permease TK2208

PF0503 putative iron ABC transporter TK2209

PF0909 ferric enterobactin transport ATP-binding protein TK0708/TK2020

PF0910 iron (III) ABC transporter, permease protein TK0707/TK2019

PF0911 iron (III) ABC transporter TK0706/TK2018

PF1519 ABC-type Fe3+ transport system, ATP-binding protein TK0158/TK0572

PF1520 ABC-type Fe3+ transport system, permease TK0159/TK0571

PF1774 iron III ABC transporter, ATP-binding protein TK0865/TK0706/TK2018 a Adjacent genes in bold are predicted to be in the same operon. b Annotation based on TIGR and NCBI databases. c Homologs in Thermococcus kodakarensis (Tk) found using BLAST. *Genes significantly up-regulated in P. furiosus iron-limitation microarray (Angeli Menon, unpublished).

CHAPTER 2

MATERIALS AND METHODS

2.1 Protein expression and purification

The vectors containing the PF1194 sequence encoding Fur (genome coordinates 1138175 to 1138558), PF0851 sequence predicted by NCBI to encode DtxR (genome coordinates 824720 to 825121), and PF0851 sequence containing the revised translation start of DtxR (genome coordinates 824684 to 825121), were constructed using the vector for protein expression, pET24dBAM, a kind gift from Chris Hopkins, University of Georgia. The pET24dBAM vector harboring the clone was a derivative of pET24d (Novagen), modified to incorporate an N- terminal hexahistidine tag (his-tag) on the expressed protein. The pET24d vector is selectable for kanamycin resistance and is designed for use in combination with a host containing a T7 lysogen under control of the lac promoter. The pET24d vector also contains a copy of the lac repressor which represses expression of the endogenous T7 RNAP except in the presence of the chemical inducer IPTG (isopropyl-β-D-thiogalactopyranoside) which causes release of the repressor from the lac promoter, thereby permitting expression of T7 RNAP. Recombinant protein expression is therefore inducible with IPTG and is driven from the T7 promoter by T7 RNAP under control of the lac promoter. For expression of the his-tagged recombinant protein, the clone was transformed into BL21-CodonPlus (DE3)-RIPL cells (Stratagene) using the manufacturer‟s protocol. Protein expression from a 1-L culture of LB (Luria-Bertani) broth (Fisher Bioreagents) was grown to an OD600 of ~ 0.8-1.0, and protein expression was induced with 1 mM IPTG. Cells were harvested 4 h after induction and resuspended in ~20 mL Binding Buffer (20 mM sodium

15 phosphate, 0.5 M NaCl, pH 7.4) containing ~10 μL of protease inhibitor cocktail (Sigma). Cells were sonicated on ice using a large horn at 40% power for 6 pulses of 30 sec each, with capping and mixing of the solution between pulses. Soluble cell extract was obtained after centrifugation for 60 min at 21,000 rpm with a Beckman JA25.5 rotor. The supernatant was centrifuged at

21,000 rpm for an additional 15 min prior to purification of the protein by column chromatography. Using the ÄKTA purifier system (Amersham Bioscience), the soluble cell extract was loaded onto a 1-mL HisTrap FF column (GE Healthcare). The column was washed with 5 mL Binding Buffer containing 20 mM imidazole followed by a gradient elution with

Eluting Buffer (20 mM sodium phosphate, 0.5 M NaCl, 0.5 M imidazole, pH 7.4) with a 25-mL

0-100% gradient. Protein-containing fractions that were relatively pure were pooled, and

Amicon® Ultra-15 centrifugal filter tube (3K NMWL, Millipore) was used for buffer exchange into 20 mM sodium phosphate, 150 mM NaCl, pH 7.4. Resulting protein was estimated to be >95% pure using SDS-PAGE. Protein concentration was determined using a Bio-Rad DC

Protein Assay kit, and aliquots of His6-Fur and His6-DtxR were stored at -80°C.

2.2 Gel filtration to determine protein quaternary structure

Gel filtration with Superdex75 HiLoad 16/60 prep grade column (GE Healthcare) was used to determine the quaternary structures of his-tagged Fur and DtxR. The column was washed with 1x column volume of buffer with low ionic strength containing 50 mM sodium phosphate,

0.05 M NaCl, pH 7.3; then equilibrated with 2x column volume of sample buffer containing 50 mM sodium phosphate, 0.15 M NaCl, pH 7.3 at a flow rate of 1.5 mL/min. 500 µL protein samples were loaded onto the column through a 500-µL sample loop according to the instruction manual. 400 µL of carbonic anhydrase (29 KDa, 3 mg/mL, Sigma-Aldrich) and albumin (66

KDa, 5 mg/mL, Sigma-Aldrich) were used as protein standards to determine the corresponding approximate molecular weight of the sample peaks.

2.3 Electrophoretic mobility shift assay

Electrophoretic Mobility Shift Assay (EMSA) was performed as a modification from that originally described in [111]. DNA probes for EMSA, representing promoter regions of selected genes, were PCR-amplified from P. furiosus genomic DNA using primers listed in Table 2.1, followed by either PCR-purification using a Qiagen PCR Purification Kit, gel-purification using a Qiagen QIAXII Gel Purification Kit, or ethanol precipitation.

Table 2.1 DNA Probes used in EMSA

Promoter Probe length Genome coordinates Forward primer Reverse primer namea (bp) ftr1* 721072-721370 cgattttcctcccatattttcac tatcgctgcttctaatgcttctc 299 feoAB* 833875-834165 actacaactcctgtctcgc caaggggattaggaataaacca 291 sipA 1872302-1872649 ggtcgtttggtgcttctac aaagagaggtgctacagtatc 348 fur* 1137971-1138289 cgttgtgaactctgctgc tgagggaaggatgagatgg 319 dtxR 824471-824748 gaagtcccgaaggagtgg acacccttgttcttttgaagg 278 aor 360149-360430 tcttagccaattcttcgtcg cgaaaatcaccgagcattg 282 suf 1214867-1215186 ccacttccattaggtccc cttggagtacccctattgtg 320 pf0849 822144-822503 tgtttttcttcatgagtgctcc tcctctctctcatcaactcc 360 a Promoter names based on the corresponding genes as annotated in NCBI and TIGR. * Probes used for both Fur and DtxR EMSA; the rest of probes were only tested with DtxR.

Incubation of DNA with various amounts of protein were carried out in 10-μL EMSA buffer (20 mM sodium phosphate, 200 mM KCl, 5% glycerol, 1 mM EDTA, pH 7.3) using a 5x stock. DNA concentration was typically 100-500 nM in each reaction, and protein was adjusted according to the molar amount of DNA. The EMSA reactions were assembled as follows. A master mix of water, 5x EMSA buffer and DNA was made according to the number of reactions in the experiment, then distributed into 0.5-mL microcentrifuge tubes on ice. Protein dilutions

17 were made in a final concentration of 1x EMSA buffer, and 2 μL of the appropriate protein dilution was added to each EMSA reaction (with 2 μL of 1x EMSA buffer added instead of protein for the DNA-only lane). For EMSA with divalent metals, 10x metal stock solutions were made freshly before use; the final concentration of metals in EMSA is 500 μM. In cases where an extra reagent was added to the reaction (e.g. divalent metal salts, heparin), volumes of water and/or 5x EMSA buffer were adjusted accordingly such that the final buffer concentration of each reaction was always 1x. Reactions were incubated at 60 °C for 20 min and immediately loaded onto a BioRad 5% TBE gel; 15-well Ready gels were typically run at 200 V for 20-30 min while the 26-well Criterion gels were typically run at 100 V for 60-110 min. The gel was then stained with ethidium bromide and imaged via UV transillumination.

2.4 Strains and growth conditions

P. furiosus strains used and constructed in this study are listed in Table 2.2. Media for culture growth was composed of 1x base salts, 1x trace minerals, 10 μM sodium tungstate, 0.25 mg/mL resazurin, with added cysteine at 0.5 g/L, sodium sulfide at 0.5 g/L, sodium bicarbonate at 1 g/L, and 1 mM sodium phosphate buffer (pH 6.8); for complex medium (YEM), containing combinations of 0.05% (wt/vol) yeast extract and 0.5% (wt/vol) maltose, or, for defined medium, containing 1x vitamin solution, 2x 19-amino acid solution and 0.35% (wt/vol) cellobiose. The

YEM was used as iron-rich medium in this research; the iron-limited medium was made based on the recipe of YEM, except the trace mineral stock was made without iron and all containers for YEM-Fe were acid-washed overnight in 2% nitric acid solution.

Stock solutions of individual components were prepared as described [112]. The 200x vitamin stock solution contained per liter, 10 mg each of niacin, pantothenate, lipoic acid, p- aminobenzoic acid, thiamine (B1), riboflavin (B2), pyridoxine (B6) and cobalamin (B12); and 4

18 mg each of biotin and folic acid. The 25x 19-amino acid stock solution contained per liter, 3.125 g each of arginine and proline; 1.25 g each of aspartic acid, glutamine and valine; 5.0 g each of glutamic acid and glycine; 2.5 g each of asparagine, histidine, isoleucine, leucine, lysine and threonine; 1.875 g each of alanine, methionine, phenylalanine, serine and tryptophan; and 0.3 g tyrosine. An additional 0.5 g/L cysteine was added to the defined medium since the amino acid stock solution lacked cysteine. Liquid cultures were inoculated with a 1-2% inoculum or with a single colony and then incubated at 90 ºC in anaerobic culture bottles or Hungate tubes degassed with three cycles of vacuum and argon. A solid medium was prepared by mixing an equal volume of liquid medium at 2x concentration with 1% (wt/vol) phytagel (Sigma) previously autoclaved to solubilize, and both solutions were maintained at 95 ºC just prior to mixing. The medium was poured into glass Petri dishes immediately after mixing. After inoculation, plates were inverted and placed into modified paint tanks, which were degassed with three cycles of vacuum and argon and incubated at 90 ºC for 48 to 64 h.

2.5 Mutant strain construction

Vectors and PCR products were prepared and transformed into COM1 strain as described previously [113]. Deletions of fur and ftr1 (PF1194 and PF0723) were constructed with 3 kb flanking regions cloned sequentially into a plasmid respectively, and this plasmid was then used as template to amplify the fur or ftr1 deletion construct with only 1 kb flanking regions and cloned into pGLW015 (from Dr. Gina Lipscomb, University of Georgia), containing the

PgdhpyrF cassette for prototrophic selection. For natural transformation, aliquots of COM1 culture typically grown to mid-log phase (~2 x 108 cells/mL) in defined liquid medium were mixed with DNA at a concentration of 2 to 10 ng DNA per μL culture, spread in 30 μL aliquots onto defined solid medium lacking uracil, and plates were placed inverted in anaerobic jars and

19 incubated at 90 ºC for ~64 h. Colonies were picked into 4-6 mL of liquid medium in Hungate tubes, on defined medium lacking uracil, and incubated anaerobically overnight at 90 ºC . For 5-

FOA selection, 30 μL of culture from Hungate tube was plated directly onto complex medium plates containing 8 mM 5-FOA and 20 μM uracil. Colonies resistant to 5-FOA were cultured in non-selective complex medium for genomic DNA isolation and screening. For genomic DNA isolation, cells from 1 mL of overnight P. furiosus culture were harvested and suspended in 100

μL buffer A (25% sucrose, 50 mM Tris-HCl, 40 mM EDTA pH 7.4) followed by addition of 250

μL 6 M guanidinium HCl, 20 mM Tris, pH 8.5, with incubation at 70 ºC for 5 min. Genomic

DNA was extracted with phenol:chloroform:isoamyl alcohol (25:24:1, buffered at pH 8), ethanol precipitated, and suspended in 50 μL 10 mM Tris buffer, pH 8.0. After PCR-confirmation of a deletion, the resulting strains were passaged twice on solid medium for colony purification. The genomic region containing the deletion was sequenced to make sure the correct genomic region was deleted. RNA was isolated from the mutant strains, quantitative PCR was then performed to confirm that no transcripts of the deleted genes were produced in the deletion mutants.

The deletion of dtxR (PF0851) was constructed by overlapping PCR with 1 kb flanking regions on either side of the PgdhpyrF cassette. Aliquots of COM1 culture grown to mid-log phase (~2 x 108 cells/mL) in defined liquid medium were mixed with DNA at a concentration of

2 to 10 ng DNA per μL culture, spread in 30 μL aliquots onto defined solid medium lacking uracil, and plates were placed inverted in anaerobic jars and incubated at 90 ºC for ~64 h.

Colonies were picked into 4-6 mL of liquid medium in Hungate tubes, on defined medium without uracil, and incubated anaerobically overnight at 90 ºC for genomic DNA isolation and screening. After PCR-confirmation of a deletion, the resulting strains were passaged twice on

20 solid medium for colony purification. Sequencing for the deletion mutant and quantitative PCR were performed to confirm the purity of the ΔDTXR strain.

Table 2.2 P. furiosus strains used and constructed in this study

Strain Genotype Parent Genome regions deleted from designation strain parent straina DSM3638 wild type - - COM1 ΔpyrF wild type PF1114 (1062504-1063123) COM1C2 ΔpyrF::pyrF COM1 None, restored FUR* ΔpyrF Δfur COM1 PF1194 (1138175-1138558) DTXR* ΔpyrF ΔdtxR::pyrF COM1 PF0851 (824642-825058) FTR1* ΔpyrF Δftr1 COM1 PF0723 (721320-722156) a Locus tags of deleted genes are listed, followed by nucleotides of the deleted region in parentheses (P. furiosus DSM 3638 genome, GenBank: AE009950.1). * Strains constructed in this study.

2.6 Growth characterization of cultures

Growth curves of FUR and FTR1 mutant were measured along with parent strain

COM1 and growth curves of DTXR strain was measured along with control strain COM1C2, which was constructed using COM1 and supplemented with pyrF (PF1114). 1-mL samples were taken hourly from growing strains in 80-mL iron-rich (YEM) or iron-limited (YEM-Fe), and centrifuged at 14,000 rpm for 10 min. Pellets were suspended in 100 μL "millipure" water and shaken gently at room temperature for 10 min. Supernatants were collected for the measurement of protein concentration, taken to represent cell protein. Bradford assay (Bio-Rad Protein Assay) was used for the measurement and serial dilutions of 10 g/L BSA protein (NEB) used as protein standard. For growth represented by culture cell densities, 200-μL samples taken from cultures were directly distributed into microtiter plates (Costar) and culture densities were measured immediately using light spectrophotometer at wavelength 660 nm.

2.7 Quantitative PCR

Cultures were grown in iron-rich and iron-limited media for total RNA isolation. Pellets were lysed in Lysis Buffer (4 M guanidine thiocyanate, 0.83% N-lauryl sarcosine, pH 5, 100 mM

β-mercaptoethanol) and 3.0 M Na-acetate pH 5.2. RNA was twice purified from cell extracts using acidic phenol-chloroform (5:1), pH 4.7, and ethanol precipitated. RNA concentration was determined using Nanodrop 2000 (Thermo scientific). RNA samples were treated with

TURBOTM DNase (Ambion) to eliminate DNA contaminants. RNA was reverse-transcribed to cDNA using the AffinityScript qPCR cDNA Synthesis Kit (Agilent Technologies). Quantitative

PCR (qPCR) reactions containing 10-ng cDNA and 1 μM qPCR primer mix according to the

Agilent instruction manual were carried out in technical triplicate using an Mx3000P instrument

(Agilent) and the Brilliant SYBR green QPCR master mix (Agilent). The constitutively expressed gene encoding the POR gamma subunit (PF0971) was selected as a control. Probes and their corresponding primers used in qPCR are listed in Table 2.3. The comparative cycle threshold method was used to analyze the resulting data, which are expressed as a ratio of gene expression change (n-fold).

2.8 Microarray transcriptomics and data analysis

Strains were grown in batch mode in a 500-mL culture bottle at 98 °C in iron-rich (YEM) and iron-limited (YEM-Fe) medium. Cells were harvested at exponential phase, immediately cooled in ice slush and centrifuged at 5,000 rpm for 15 min. Total RNA was isolated and prepared as described in section 2.7 and then was reverse-transcribed to cDNA using the

AffinityScript qPCR cDNA Synthesis Kit (Agilent Technologies). RNA was hydrolyzed using

1M NaOH for the isolation of cDNA. cDNA was extracted using phenol-chloroform-AA

(25:24:1), pH 8.0, and quality checked using Nanodrop 2000 (Thermo scientific). Finally, ~1 μg

22 purified cDNA from each condition was labeled differentially with Alexa dyes 546, 594, or 647

(Molecular Probes, Eugene, OR) according to the manufacturer‟s instructions and purified using

Micro Bio-Spin Chromatograpy Columns (BIO-RAD). Labeled cDNA pools derived from P. furiosus cells grown in iron-rich or iron-limited medium were combined and hybridized to the microarrays by using a Genetac hybridization station (Genomic Solutions, Ann Arbor, MI) for

16 h. The slides were then washed automatically for 20 s in each of 2x SSC (1x SSC is 0.15 M

NaCl plus 0.015 M sodium citrate) with 0.1% Tween 20, 0.2x SSC with 0.1% Tween 20, 0.2x

SSC, and finally rinsed in distilled water and blown dry with compressed air. Fluorescence intensities of each of the three dyes were measured with a Scan Array 5000 slide reader (Perkin-

Elmer) with the appropriate laser and filter settings.

Fluorescent spots on the microarrays were identified and quantified by using the Gleams software package (Nutec, Houston, TX). The relative amounts of the transcripts were presented in a linear fashion by using log2(ratios) (iron-limited/iron-rich). The detection limit of fluorescent signals was set arbitrarily to 500 intensity units, and such spots are not visible on the false overlay. Only ORFs that display intensities of more than the detection limit were considered valid. For the COM1C2 microarray, each log2(ratio) value represents an average of two hybridization experiments by using cDNA derived from three different cultures: two grown in iron-limited and one grown in iron-rich medium. For the DTXR, each log2(ratio) value represents an average of four hybridization experiments by using cDNA derived from four different cultures: two grown in iron-limited and two grown in iron-rich medium. Standard deviations for these data are included.

2.9 Determination of intracellular iron content

500 mL of fresh iron-rich (YEM) and iron-limited (YEM-Fe) cultures were inoculated with mid-log phase cultures of COM1 and FTR1. Cells were harvested at exponential or stationary phase. Cells were centrifuged at 15,000 rpm for 10 min, and the pellets were washed three times with 1x base salt solution to remove extracellular traces of salts. Lysis Buffer containing 20 mM sodium phosphate, pH 7.4, made with ultrapure water was added at 3mL lysis buffer per ~1 g cells, and cells were incubated at room temperature with vigorous shaking for 30 min. Lysates were then ultracentrifuged at 30,000 rpm for 1 h and supernatant was collected for further analysis. Intracellular iron contents were analyzed by inductively coupled plasma mass spectrometry (ICP-MS) using Agilent 7500ce ICP-MS. The iron contents measured by ICP-MS were normalized to the protein concentration of each sample for the analysis.

Table 2.3 Probes used in Quantitative PCR

Probe Genome Probe Forward primer Reverse primer name a coordinates length (bp) ftr1* 721498-721662 acgggggaattgaagagaaggagc accccactaaggcaagagga 165 feoAB* 833317-833508 accgcccatagcattgatgagt agcccctttctttttgagtagatca 192 pf0911 883501-883666 ccggcccaggatgcctaaga ggaagctttccagggccacc 166 pf0503 522196-522022 cgggagttgccgttgctgag actggagccgttatcgtcctgg 175 pf1774 1649464-1649627 tctgctcatagtgcttgtgtgga tctggggagtcattgtttgggt 164 fur 1138190-1138319 gccgttagagtactgaaggagaaggg actcttctttgagccttttgaaaacc 130 dtxR 824788-824985 acccccgaccgttgttgagg tggcacgcatctctctcagc 198 aor 359880-360059 atcatcgccgctggtcctct gggcttctccgccttaccct 180 sufC 1214762-1214903 acgaactccacgtagtaatgggacc tcctctggaggaagcgacgtt 142 a Gene names as annotated in NCBI and TIGR. * Transcriptional response determined for both FUR and DTXR, while the transcription of the rest were only determined for FUR.

CHAPTER 3

CHARACTERIZATION OF THE PUTATIVE IRON-RESPONSIVE TRANSCRIPTION

FACTOR FUR

3.1 Introduction to the Fur family proteins

The Fur (ferric uptake regulator) family of proteins is widely distributed among bacteria and known to regulate gene transcription globally (Fig. 3.1). It was initially described as an iron- responsive repressor of the iron-transport systems in Escherichia coli [114], hence its name. It has also been found in several other organisms that genes encoding iron transporters are up- regulated by Fur in Fe-limited conditions, while the expression of genes encoding oxidative stress enzymes, iron storage and non-essential Fe-containing proteins was decreased [115-118].

The general working model for Fur posits that the coordination of one Fe2+ per monomer enables the dimeric protein to bind a specific 19-bp DNA sequence, called the “Fur box”

(GATAATGATwATCATTATC, w = A or T), within the promoter of the regulated genes. The affinity of Fur for Fe is poised to allow accumulation of sufficient intracellular iron to activate essential iron-containing and iron-utilizing enzymes. However, when iron levels exceed those needed for metalloenzyme function, Fur represses further uptake and thereby helps prevent iron overload. Typically, the binding of Fe2+-Fur on DNA hinders the access of RNA polymerase resulting in the repression of downstream genes (mechanism 1, Fig. 3.1). Fur has been implicated as both a repressor and activator of gene expression. Genes that require Fur and iron for their efficient expression include iron sodB (superoxide dismutase), sdhCDAB (succinate dehydrogenase), acnA (aconitase A), fumA (fumarase A) and ftn (ferritin) [119]. The positive

25 regulation by Fe2+-Fur is often indirect (mechanism 2b, Fig. 3.1), mediated by Fur-dependent repression of an anti-sense regulatory small RNA (sRNA), such as RyhB. This RNA acts at the post-transcriptional level to repress a large number of genes encoding iron-using proteins, resulting in redistribution of the intracellular iron in different organisms, such as E. coli, P. aeruginosa, V. cholerae, S. flexneri, and cyanobacteria [120]. Other iron-responsive sRNAs include Fur-regulated NrrF in N. meningitidis, which facilitates the degradation of sdhA and sdhC mRNA [121] and PrrF1 and PrrF2 from P. aeruginosa that lead to the rapid loss of mRNAs for sodB, sdh and a gene encoding a bacterioferritin [122], resulting in indirect activation of genes by Fur. It has also been suggested that some Fur proteins can act as a direct transcriptional activator (mechanism 2a, Fig. 3.1). In N. meningitidis, Fe2+-Fur binds upstream of the promoter of norB and directly activates transcription by RNA polymerase in vitro [123]. Similarly, iron- bound Fur activates the transcription of nifS in H. pylori [124].

In the above cases, where Fur acts as either a repressor or activator for gene expression, the DNA-binding of Fur requires bound iron. Evidence for Fur DNA binding in the absence of iron has been most clearly documented in H. pylori. In this organism, the type of classical regulation by Fe2+-Fur is used to control expression of several genes including the aliphatic amidase, amiE, which plays an important role in ammonia production through the hydrolysis of aliphatic amides [118, 125, 126]; meanwhile, Fur regulation in H. pylori goes beyond the classic paradigm. In fact, Fur has also been shown to repress expression of some promoters in an iron- depleted (apo) form (mechanism 4, Fig. 3.1) [127-129]. For this apo-regulation, in the absence of iron the apo-Fur protein can bind to the promoters of its target genes and block transcription.

Thus, genes repressed by apo-Fur are transcribed in iron-replete conditions. Currently, the apo-

Fur regulon is predicted to contain 16 gene targets [125, 126]. Of these targets, only sodB,

26 encoding the superoxide dismutase important for oxidative defense, and pfr, encoding the iron storage molecule, have been definitively shown to be directly regulated by apo-Fur [127, 129,

130]. Expression of both of these genes is repressed by apo-Fur when iron is limited, but this repression is lost in a fur mutant strain. Just like Fe2+-Fur, apo-Fur may also play a direct positive regulatory role (mechanism 3, Fig. 3.1). In V. vulnificus, the Fur protein activates its own expression, and the extent of its activation is elevated in the presence of iron chelator added to the growth medium. To verify that the positive regulation is not indirect through sRNA, the homolog of the sRNA-binding protein Hfq was deleted, resulting in no obvious alteration in the amount of Fur protein detected in the deletion mutant in comparison with that of wild-type V. vulnificus, and suggesting that positive auto-regulation of fur expression is not mediated by Hfq

[131]. The Fur-family iron-responsive regulator, Irr, was also found to control positively the transcription of ferric siderophore receptor genes in B. japonicum [132].

Fur family proteins are widespread within the Bacteria and members of Fur homologs characterized include Zur (zinc uptake regulator) [133, 134], Mur (manganese uptake regulator)

[135, 136], Nur (nickel uptake regulator) [137], and the peroxide-sensing protein PerR [138-140].

Although these proteins show a wide diversity of metal selectivity and biological functions, they share a similar architecture comprising two domains: an N-terminal DNA-binding domain containing a winged-helix motif and a C-terminal dimerization domain, which are connected by a „hinge‟ loop. Structural information is available for a few full-length Fur proteins, the

Pseudomonas aeruginosa PaFur [141], the Vibrio cholerae VcFur [142], the Helibactor pylori

HpFur [143] and the DNA-binding domain of E. coli EcFur [144]. Crystal structures of other

Fur-like proteins have also been described with those of the Mycobacterium tuberculosis MtZur

[145], the Bacillus subtilis peroxide-regulon repressor BsPerR [146-148], and the Streptomyces

27 coelicolor ScNur [149] confirming the structural homology between the members of this family.

Despite the same secondary structure and overall architecture shared by all Fur-family proteins, the metal-binding sites in different family members can be diverse in respect to numbers, composition, and position of metal binding residues in the proteins. In Pseudomonas aeruginosa, the PaFur crystal structure contains two Zn2+ binding sites, site 1 and 2; site 1 is found in the C- terminal dimerization domain and is hexacoordinated by two histidines, a glutamic acid, a bidentate aspartic acid and a water molecule; site 2 is located in the hinge region between the

DNA-binding domain and dimerization domain and exhibits a tetrahedral geometry with two histidines and two glutamates as ligands [141]. It is not known that Zn2+ is the naturally occurring metal. In Bacillus subtilis PerR, four cysteine residues (C96, C99, C136 and C139)

2+ cluster to form a high-affinity Zn -binding site (Cys4Zn) not found in PaFur; five other residues

(H37, D85, H91, H93 and D104) are candidate ligands for the regulatory metal ion, Fe2+ or Mn2+

[138, 147]. Three distinct Zn2+-binding sites were found in MtZur; two of the metal ions are in the dimerization domain, and one is located in the hinge region between the dimerization and the

DNA binding domain. Site 1 is surrounded by Asp62 and Cys76 from the DNA binding domain and His81, His83 from the dimerization domain; site 2 is tetrahedrally coordinated to a cluster of sulfur ligands: Cys86 and Cys89 from the dimerization domain and Cys126 and Cys129 located at the C terminus; site 3 is located in the core of the dimerization domain where the Zn2+ ion is tetrahedrally coordinated by three histidines (His80, His82 and His118) and one glutamate

(Glu101) [145]. Sequence alignment of the structurally characterized Fur-family members illustrates that the metal-binding sites in Fur are conserved through species, found in PaFur,

HpFur, and VcFur (except an additional site 3 in HpFur); however, different metal-binding

28 residues are found in ScNur, MtZur and BsPerR (Fig. 3.1), implying different metal-specificities for these regulators.

3.2 Fur homolog in Pyrococcus furiosus

P. furiosus requires iron and can survive in conditions of different iron concentration. It may be capable of facing changes affecting its internal iron balance, and therefore needs a regulatory system. The genomic sequence of P. furiosus predicts a putative fur homolog gene,

PF1194. Analysis of PF1194 protein sequence revealed homology with the conserved protein domain of the Fur family of transcription factors (Fig. 3.2) [150]. PF1194-encoded protein

(PfFur) shares 32% identity and 58% similarity with E. coli Fur, 35% identity and 61% similarity with Bacillus subtilis PerR, 22% identity and 46% similarity with Mycobacterium tuberculosis

Zur and 24% identity and 47% similarity with Streptomyces coelicolor ScNur. The sequence alignment of PfFur with PaFur, BsPerR, MtZur and ScNur revealed that residues from the metal- binding site 2 of PaFur are conserved in PfFur. This site is located between the DNA-binding domain and dimerization domain and predicted to be essential for the protein to function as a regulator. The metal-binding site 3 in the dimerization domain of PaFur is absent from PfFur, as are the sites consisting of two CXXC motifs in MtZur and BsPerR, which were proven to be essential for regulatory activity (Fig. 3.3). PfFur protein shared the strongest sequence similarities (85%) with TK1058-57 in T. kodakarensis KW128, a -1 frame-shifted fur homolog consisting of two distinct reading frames, TK1058 and TK1057; the “corrected” Fur gene in T. kodakarensis seemed to be inactivated for iron metabolism, as neither its expression nor its protein's activity was regulated by iron [151]. Other close hits (with e-values less than 10-45) include Thermococcus barophilus MP (60% sequence identity) and Pyrococcus sp. NA2 (56% sequence identity).

3.3 DNA-binding affinity of PfFur

3.3.1 Expression and purification of recombinant his-tagged PfFur

To test whether PfFur was a sequence-specific DNA-binding protein, the protein was overexpressed as a his-tagged recombinant protein in E. coli (as described in Section 2.1) for in vitro studies. The recombinant protein was produced using a protein-expression vector which allowed for the expression of PfFur with an N-terminal his6-tag which facilitates protein purification. Native PfFur is 127 amino acids in length and has a calculated molecular weight of

14,842.55 Da; the recombinant PfFur with the addition of a his6-tag produces the construct Met-

Ala-His6-Gly-Ser-protein, and the new molecular weight is 15,880.60 Da. The BL21-CodonPlus

(DE3)-RIPL strain that compensates for rare codons by expressing supplementary corresponding tRNAs (for Arg, Ile, Pro, and Leu rare codons) was utilized for efficient expression of PfFur in E. coli host strain. Expression of the protein did not appear to be toxic to the cells and did not significantly slow culture growth; moreover, the protein was expressed in abundance, predominantly in soluble form, which greatly simplified the purification procedure (Fig 3.4).

The his6-tagged protein was purified from cell extract using a nickel-affinity column, and a gradient of imidazole was used to elute the protein from the column. Only the fractions which were of acceptable purity were collected and pooled for buffer exchange using a centrifugal filter tube. An example gel showing protein purified in this manner can be seen in Figure 3.4. Once the protein was exchanged into a suitable salt-containing buffer, small aliquots were stored at -80 °C to be thawed individually for use in in vitro assays.

3.3.2 PfFur neither recognizes DNA probes nor responds to iron

In E. coli, both the transcription of the fur gene and the activity of the Fur protein are driven by iron availability [152, 153]. The Fe2+-Fur complex can bind to specific sequences in

30 promoters. In order to verify the binding ability of his6-tagged Fur to DNA, its own promoter region, containing ~200-bp upstream and ~50-bp downstream of the Fur ORF start, was chosen as a probe and electrophoretic mobility shift assays (EMSA) were performed to determine the ability of recombinant Fur protein to bind to DNA. No binding-shift was observed in the gel, which indicates no stable complex was formed between the recombinant Fur and its own gene promoter. Sensitivity of Fur protein to divalent metal ions was measured with transition metals

Fe2+, Mn2+ or Zn2+ using EMSA; at high protein/DNA ratio (16- to 32-fold protein to probe), the binding of Fur to DNA probe was increased in the presence of transition metal ions, however, the improved binding seemed to be non-specific since the binding-shift always appeared to be a smear in the gel (Fig. 3.5). Different DNA probes were then designed to confirm this result. The promoters of ftr1 and feoAB genes, both containing ~200-bp upstream and ~50-bp downstream of the ORF start, were chosen based on the results of iron-limitation microarray transcriptomics

(Angeli Menon, unpublished data) in which the transcription of both operons were regulated in response to iron (Fig. 3.6). No binding complex was formed between Fur protein and promoters of ftr1 and feoAB genes, either in the presence or absence of transition metal ions. Finally, to broaden the range of DNA probes used for the detection of PfFur binding ability, an artificial

DNA library was constructed. The designed single-strand probes consist of a 30-nt randomized region flanked by primer regions containing EcoRI restriction sites (Table 3.1). Double-strand

DNA library was PCR-amplified using the synthetic single-strand probes with primers. To create the dsDNA probe, 100 pmol of single-stranded DNA library probe was amplified with 2 nmol of each primer for a total of 5 PCR cycles. The PCR-amplified double-stranded DNA library was polyacrylamide gel-purified for EMSA. EMSA of his6-tagged Fur with the artificial DNA library was performed alongside SurR (S0-responsive regulator) protein (a kind gift from Dr. Gina

Lipscomb), which has been previously shown to be a specific DNA-binding transcription factor

[154], as a positive control (Fig. 3.7); while binding-shifts were observed for SurR with the DNA library (bindings started at protein/DNA ratio ~4); for Fur, at high protein/DNA ratio, the amount of template library DNA in the gel was decreased, yet there were no shifted bands observed, indicating that the disappearance of DNA in the gel was caused by non-specific binding.

Table 3.1 DNA probe and primers for artificial DNA library

Name Sequence a

DNA library single-stranded probe ggtctagagaattcaagcttc(n)30ggatccgaattcgtcgac Library primer F gctcaggtctagagaattcaa Library primer R actactgtcgacgaattcgga a EcoRI sites used in cloning are colored red.

3.4 Characterization of the fur deletion mutant

3.4.1 Validation of Pyrococcus furiosus fur deletion strain

To further investigate the effect of Fur and its response to iron, a Pf strain containing a markerless deletion of fur (designated FUR) was constructed in COM1 background strain

(ΔpyrF); this was supplied by Dr. Karen Stirrett, University of Georgia. It was constructed by cloning the 3 kb flanking regions of fur into a plasmid, and this plasmid was then used as template to amplify the fur deletion construct with only 1 kb flanking regions and cloned into pGLW015 (from Dr. Gina Lipscomb, University of Georgia), containing the PgdhpyrF cassette for prototrophic selection. The vector was thus mixed with freshly prepared COM1 culture and plate on solid medium without uracil to let the transformation begin. The first round selection was for crossover event 1, in which the vector containing the PgdhpyrF cassette was integrated into COM1 genome and made the resulting strain uracil prototrophic. Colonies were isolated, grown and plated for the selection of crossover event 2, in which the vector was removed from

32 the genome by homologous recombination and left a fifty-percent chance for the generation of fur deletion strain (Fig. 3.8). PCR and sequence analyses were used to confirm gene deletion in

FUR. mRNA was harvested from wild type and FUR strain for qPCR; the result indicated that fur was detected for wild type around 16th cycle of PCR, while in FUR the amplification curve only started to increase after 35th cycle possibly due to non-specific amplification of PCR products (Fig. 3.8). Hence, it was concluded that FUR is a clean deletion.

3.4.2 Effect of iron availability on FUR growth

Growth of FUR along with COM1 and wild type strains were compared after three consecutive transfers with inocula at mid-exponential phase (~5 x 107 cells/mL) in an iron- limited maltose-based medium. For FUR and COM1, 40 μM uracil was supplemented to the medium to assure growth.

Growth of the FUR and COM1 was studied in 10 mL iron-rich media containing 7.6

μM Fe, iron-limited media in which no iron was added (Fe concentration is < 0.8 μM as measured by ICP-MS), and the iron-limited media containing 10 μM, 50 μM and 100 μM of the

Fe2+-specific chelator bathophenanthroline disulfonate (BPS). Samples were taken from the

COM1 and FUR cultures in early stationary phase at 11 h and the amount of cytoplasmic protein was measured and compared (Fig. 3.9) between the two strains. FUR had generally the same response to iron as COM1; since growth of the two strains started to be impaired dramatically at 50 μM BPS in media, it was impossible to use for growth study; the iron-rich containing 7.6 μM Fe and iron-limited medium with no iron added were chosen for further growth study. As expected, the growth of FUR was similar to COM1 and wild type in iron-rich

33 media; in iron-limited media, growth of the three strains were similar still (Fig. 3.9), even though the growth of all three in iron-limited media were impaired compared to that in iron-rich media.

3.4.3 Transcriptional response in FUR

Previous studies showed that the growth of Pyrococcus furiosus in iron-limited conditions is impaired and transcription of a number of iron-related genes, including iron transporters and genes encoding iron-containing proteins, was regulated in response to iron.

Table 3.2 provides a list of some of the iron-regulated genes identified in previous iron-limitation microarray studies (Angeli Menon, unpublished data). Among all genes regulated by iron, the ftr1 (PF0723) gene appeared to be most sensitive to iron, the transcription of which was up- regulated more than 7-fold in iron-limited condition. ftr1 encodes a high-affinity iron transporter, it shares 38% identity and 56% similarity with its E. coli homolog EfeU, a ferrous iron permease.

A putative bacterial Fur-box was located upstream of the efeU gene and the expression of EfeU in E. coli has been proved to be regulated by Fur in response to iron [155]; therefore, the transcriptional response to Fur of PfFtr1 was selected to study in our deletion mutant.

Two other genes involved in putative iron-uptake systems, feoB (PF0857) and PF0911

(ABC-type iron transporter), were also found to be regulated by iron. feoB is in an operon with upstream gene feoA (PF0858), encoding a high-affinity ferrous iron uptake system, FeoAB, which is well conserved in bacteria and plays an important role in bacterial iron homeostasis.

PF0911 is in an operon containing PF0911-0909, encoding an iron ABC-transport system consisting of a periplasmic siderophore-binding protein, a transmembrane permease, and an

ATPase. PF1285 and PF1286 are annotated as SufD and SufB, homologs of which are essential for FeS cluster assembly in bacteria. PF0346 encodes AOR, which has been characterized as an iron- and tungsten-containing protein [75] and PF0851 is homologous to DtxR, another iron-

34 responsive regulator that will be discussed in detail in the next chapter. In previous research, the bacterial homologs of all of the genes in Table 3.2 have been shown to be under the control of

Fur in bacteria [114, 156], and thus they are hypothesized as Fur-regulated genes in P. furiosus and as targets of our transcription study in FUR strain. Besides these genes, two other genes annotated as putative iron ABC-transporters, PF0503 and PF1774, were added to our list for consideration. PF0503 composes an operon with PF0502, an iron(III) dicitrate transport system permease protein fece; while PF1774 is predicted to be an ATP-binding protein of the iron(III)

ABC transport system. PF0503-0502, PF1774, along with the PF0911-0909 operon, represent the putative iron ABC transport systems predicted by the NCBI database.

Table 3.2 Transcription response to iron-limitation in P. furiosusa

ORF Gene nameb ORF description / operon funcationb Fold change from +Fe to –Fe Up-regulated >2-fold without Fe PF0723 ftr1 Iron permease FTR1 family 7.6 PF0857 feoB ferrous iron transport protein b 4.7 PF0911 iron(III) ABC transporter 2.1 PF1285 sufD FeS assembly protein SufD 5.1 PF1286 sufB cysteine desulfurase activator SufB 6.2 Down-regulated >2-fold without Fe PF0346 aor aldehyde:ferredoxin oxidoreductase 1/2.5 PF0851 dtxR iron-dependent repressor 1/2.0 a Derived from Fe-limitation microarray data from Angeli Menon‟s unpublished work. b Gene names and descriptions are from TIGR and NCBI databases.

Cultures of COM1 and FUR were cultivated in iron-rich and iron-limited media as defined previously and harvested in the mid-exponential phase, about an hour before the growth entering stationary phase in the iron-limited media. Transcriptional responses of the genes potentially regulated by Fur were determined using qPCR. The qPCR results were given as fold

35 regulation of selected ORFs comparing DTXR to COM1C2 strain. As shown in Figure 3.10, compared to COM1, most of the iron-responsive genes were regulated at a similar level in FUR.

Overall, since: (1) PfFur does not specifically bind to DNA, especially promoters of predicted fur regulon genes; (2) FUR shows no growth phenotype; and (3) the transcriptional regulation of genes, including iron transporters, was the same in FUR as it was in COM1, it is concluded that Fur does not likely play an important role in iron-responsive regulation in P. furiosus.

Figure 3.1 Regulatory potential of bacterial Fur proteins. A. Tree-view of the occurrence of the fur homologs in bacteria and archaea, generated using STRING 9.0 (http://string.embl.de). B.

Fur proteins have been described to regulate gene expression by four general mechanisms. In most cases, it is the iron-loaded form of Fur protein (top) that binds to DNA operator sites.

DNA-bound Fur can directly repress transcription (mechanism 1) or activate gene expression either directly or indirectly via small RNA molecules (mechanisms 2a and 2b, respectively). In H. pylori, the iron-free (apo) form of Fur also binds DNA and represses expression of an iron storage protein (mechanism 4). In the case of B. japonicum Irr, apo-protein binds DNA in the absence of a metal ion cofactor and can either activate (mechanism 3) or repress (mechanism 4) the expression of target genes. In general, when Fur functions as a repressor, the Fur box overlaps the promoter region, whereas activation is associated with a Fur box just upstream of the promoter [106]. C. Sequence alignment of Fur and Fur-like proteins. The amino acid sequences of Fur proteins from H. pylori (HpFur), V. cholerae (VcFur), P. aeruginosa (PaFur), E. coli (EcFur) and of the Fur-like Zur from M. tuberculosis (MtZur), Nur from S. coelicolor (Nur) and B. subtilis PerR (PerR) were aligned using CLUSTALW2. Secondary structure elements are indicated above HpFur sequence; the DNA-binding domain is shown in green, the dimerization domain in violet, the connecting loop in beige and helix α6 in cyan. Residues involved in metal- binding site 1 (or equivalent) are shaded in cyan, site 2 in red, site 3 in yellow [143].

Figure 3.2 PfFur (PF1194) protein sequence analysis. A. Conserved domain search results for

Fur protein sequence from online tools available at NCBI. PfFur sequence is represented by a line segment colored in grey and blue with matching conserved domains indicated below.

Conserved domain descriptions and sequence alignments are shown. For the sequence alignments, identical residues are colored red and similar residues are colored blue. B. PfFur protein secondary structure predicted using PSIPRED [41]. Helices are indicated as purple rods and strands are indicated as yellow arrows.

CD Length: 116 Bit Score: 67.60 E-value: 6.45e-13

gi 18893279 19 PQRLKMLEVIEELGpSHPSLNEVFKRLKEEFPTLSFSTLYSNVMTLKELGLIETLPINDE-TRI 81 Cdd:cd07153 1 PQRLAILEVLLESD-GHLTAEEIYERLRKKGPSISLATVYRTLELLEEAGLVREIELGDGkARY 63

gi 18893279 82 EINTKPH 88 Cdd:cd07153 64 ELNTDEH 70

Figure 3.3 PfFur protein sequence compared to orthologs in other species. A. Protein sequence alignment of PfFur with PaFur, BsPerR, MtZur and ScNur using CLUSTALW2.

Residues involved in metal-binding site 1 (or equivalent) are shaded in cyan, site 2 in red, site 3 in green. Conserved residues in PfFur are shaded in grey. B. Fur homologs conserved in

Thermococcaceae species and sequence identities with PfFur. The Fur homolog in T. kodakarensis is cryptic with -1 frameshift between TK1057 and TK1058.

Figure 3.4 His6-tagged PfFur protein purification. A. SDS-PAGE of PfFur expressed in E. coli BL21. Lane 1: before induction, lane 2: after induction (4 h) with IPTG. B. Results for automated chromatographic purification of his6-PfFur from soluble cell extract using a nickel- affinity column with Binding Buffer (20 mM sodium phosphate pH 7.4, 0.5 M NaCl, 20 mM imidazole) and Eluting Buffer (20 mM sodium phosphate pH 7.4, 0.5 M NaCl, 0.5 M imidazole).

Protein was eluted with 25-mL 0-100% gradient of Eluting Buffer. PfFur elution peak is indicated by arrow. C. SDS-PAGE gel of fractions collected from elution step of nickel-affinity chromatography. The fractions were pooled for buffer-exchange into 20 mM sodium phosphate,

150 mM NaCl, pH 7.4.

A C

Figure 3.5 EMSA of PfFur with its own promoter. A. 100nM PF1194 promoter was incubated with PfFur at corresponding Protein/Nucleic acid ratios listed above each lane. B. Effect of divalent metals (Fe2+, Mn2+ and Zn2+) on PfFur DNA-binding ability; experimental parameters remained the same except 1μL of 5 mM metal stock solution added to each assay and the amount of 5x EMSA buffer was changed correspondingly.

A B

Figure 3.6 EMSA of PfFur with feoAB and ftr1 promoters. 100 nM promoter of (A) feoAB or

(B) ftr1 was incubated with PfFur at corresponding Protein/Nucleic acid ratios listed above each lane. Effect of Fe2+ on PfFur DNA-binding ability was tested with feoAB promoter.

A B

Figure 3.7 EMSA of PfFur with artificial DNA library. 500 nM artificial DNA library containing 30-bp random sequence was incubated with PfFur at corresponding Protein/Nucleic acid ratios listed above each lane. SurR, a S0-responsive transcription factor, was used to bind the same DNA library alongside as a positive control.

Figure 3.8 Construction and verification of the fur deletion strain. A. The knockout vector was constructed with 1 kb fur gene flanking regions and the PgdhpyrF cassette for prototrophic selection. The knockout vector was transformed to the naturally competent strain COM1 (ΔpyrF); the first crossover event resulted in the integration of the vector into COM1 genome and made the strain uracil prototrophic; in the second crossover event the vector was removed from genome DNA, and the resulting strains would be either the original COM1 or the strain with fur deletion. B. Colonies from the second crossover event were picked and genomic DNA was isolated. Primers were designed to amplify the region containing the 1kb fur flanking region.

Agarose DNA gel was performed to confirm their length, as the PCR product from the fur deletion strain should be shorter than the original sequence. All the experiments above were performed by Dr. Karen Stirrett.

Figure 3.9 Growth of COM1 and FUR in response to iron in medium. A. Medium with

BPS was made by adding corresponding amount of 100 mM BPS stock into the YEM-Fe medium. Cell extracts from 1-mL culture of COM1 and FUR were prepared with Millipore water and vigorous shaking. Protein concentrations of COM1 (black) and FUR (white) were measured using Bradford assay and are compared for a range of growth conditions as specified.

B. Growth curves of COM1 (blue) and FUR (red) grown in iron-rich (solid) and iron-limited

(empty) medium. Inocula for the growth were passed three times in YEM-Fe to remove the residue iron associated with cells; all glassware was washed with 2% nitric acid overnight.

Protein concentrations were measured for COM1 and FUR as mentioned in text.

Growth of COM1 and FUR in Media containing Range of Concentrations of Iron

45 40 COM1 35 FUR 30 25 20 15 10

5 ug Cell Cell Protein/ ug ml Culture 0 YEM YEM-Fe 1uM BPS 10uM BPS 50uM BPS 100uM BPS

Growth of FUR in Fe-rich and Fe-limit Media 60.00

COM1 -Fe 50.00 FUR -Fe 40.00 COM1 +Fe FUR +Fe 30.00

20.00

ug Cell Cell Protein/ ug ml Culture 10.00

0.00 0 2 4 6 8 10 12 14 Time (hr)

Figure 3.10 Regulation of transcription of the iron-related genes in COM1 and FUR. The regulation of transcription in response to iron was compared in COM1 and FUR. Total RNA was isolated from COM1 and FUR grown in iron-rich and iron-limited medium at exponential phase and transcribed to make cDNA; qPCR was performed with primers designed for genes including ftr1 (PF0723), feoA (PF0858), pf0911, pf1774, pf0503, sufC (PF1287), aor (PF0346), fur (PF1194) and dtxR (PF0851). "Fold changes" (y-axis) in COM1 (black) and FUR (white) were derived from the comparison of the transcription level in iron-limited to iron-rich medium.

All data were normalized to the reference gene por (PF0971).

CHAPTER 4

CHARACTERIZATION OF PUTATIVE IRON-RESPONSIVE TRANSCRIPTION

FACTOR DTXR

4.1 Introduction to the DtxR family proteins

DtxR was first characterized in Corynebacterium diphtheria, in which it mediates the induction of expression of the diphtheria toxin when iron is limiting [157, 158]; DtxR also controls iron uptake and thereby plays a role functionally similar to the ferric uptake regulator

(Fur) protein. DtxR-like proteins have also been characterized in Streptomycetes and in

Mycobacteria, for which they are called IdeR [159]. Generally, DtxR-like proteins regulate a spectrum of genes [160-163] encoding putative transporters, proteins involved in siderophore synthesis and iron storage in Gram-positive bacteria with high GC-content, in a manner similar to those regulated by Fur in Gram-negative bacteria. The basic fold of the Fur and DtxR family members consists of two domains: an N-terminal DNA-binding domain and a C-terminal domain responsible for dimerization and metal binding. The winged-helix DNA binding domain of Fur resembles structurally that of DtxR, despite a lack of significant sequence similarity [141]. The consensus DNA-binding sequence identified for DtxR also differs from the Fur box.

Comparison of protein sequences within the DtxR family shows that they have high sequence similarity and many conserved metal-binding residues (Fig. 4.1), although in some of the family members the C-terminal SH3-like domain, which may modulate metal binding and

DNA-binding affinity, is missing [164]. Despite similarity in structure and many conserved residues, different members of this family are activated by metal ions in distinct manners. Crystal

58 structures of DtxR from C. diphtheria and its closely related homolog, IdeR from M. tuberculosis, revealed two metal-binding sites in each subunit 9 Å apart (Fig. 4.1); however in vivo only Fe2+ acts as a co-repressor, while several other divalent transition metals including

Co2+, Mn2+, Cd2+, Ni2+ and Zn2+ also activate the protein. The B. subtilis manganese transport regulator MntR is the prototype for a subgroup of proteins from the DtxR/IdeR superfamily that respond to Mn2+ rather than Fe2+ [165]. It is a two-domain protein, lacking the C-terminal domain found in DtxR and IdeR. MntR binds to its operators to prevent expression of manganese transport systems encoded by the mntABCD and mntH operons with high-selectivity in vivo and in vitro for Mn2+ [107, 166-168]. This DNA binding activity is distinctly diminished by other metal cations including Mg2+, Ca2+, Fe2+, Co2+, Ni2+ and Zn2+. Crystal structure shows MntR binds two metal ions 4.4 Å apart, forming a binuclear complex at the interface of its DNA- binding and dimerization domains (Fig. 4.1). ScaR from S. gordonii, which selectively represses expression of the manganese transporter (scaCBA operon) in the presence of co-repressor Mn2+, showed a single occupied metal-binding site that is unique to ScaR [169]. Residues at the primary site in DtxR (Met10, Cys102, Glu105, and His106) that are known to be essential for

DNA binding [170] are somewhat conserved in ScaR (Asp7, Glu99, Glu102, and His103), although no metal binding was observed at that site.

The importance of the metal-binding geometry is illustrated by the modification of the

MntR metal-binding residue Asp8 to methionine. This substitution leads to a protein that binds a single metal ion per subunit and has reduced specificity for metal ion effectors and lower affinity for DNA [107, 168]. Similarly, modifications to sulfur-containing ligands in DtxR, replacing

Cys102 with glutamate or Met10 with aspartate, generate a mutant DtxR that is strongly

59 activated by Mn2+ in B. subtilis cells [107], indicating that ligand selection plays a role in allowing DtxR to discriminate against the noncognate metal ion manganese.

The characterized members of the DtxR family thus fall into two groups: those that respond to iron and those that respond to manganese. The relative occurrence of DtxR-like regulators that respond to manganese compared to those responsive to iron is unclear since the metal specificities of many members of the family have yet to be fully characterized.

DtxR family members are also widespread in archaea. The DtxR homolog, MDR1

(metal-dependent repressor) found in the archaeon Archaeoglobus fulgidus, cotranscribed with a putative ABC-type metal transporter, has been used as a model to study transcriptional regulation in archaea [45]. The expression of MDR1 is dramatically up-regulated in response to decreased iron availability in the growth medium and the binding of MDR1 to its promoter is induced in the presence of Fe2+. It was shown that MDR1 repressed transcription from its own promoter in vitro by interaction with TBP-TFB to prevent RNAP recruitment.

In Thermococcus kodakarensis, the DtxR homolog encoded by TK0107 was identified as an iron-responsive regulator. Although the recombinant DtxR protein did not show any DNA- binding activity, the deletion mutant can grow in iron-limited medium with higher concentration of the iron chelator 2,2‟-dipyridyl compared to wild type. Transcription of the two FeoAB high- affinity ferrous transport systems (TK0714-TK0716 and TK0957-TK0958) was activated in the dtxR deletion mutant when compared to wild type, indicating DtxR is a repressor for the iron uptake when environmental iron is not limiting [151].

4.2 DtxR homolog in Pyrococcus furiosus

As discussed above, P. furiosus contains a homolog of the iron-responsive transcription regulator DtxR; PfDtxR is encoded by PF0851. It has a predicted length of 133 amino acids and

60 contains sequence conserved in the DtxR family of proteins (Fig. 4.1.a) as shown by alignment with the well-characterized DtxRs of Corynebacterium diphtheria DtxR [171, 172],

Mycobacterium tuberculosis IdeR [159, 173, 174], Streptococcus gordonii ScaR [169] and

Bacillus subtilis MntR [165, 167, 175, 176]. However, the primary sequence comparison also reveals several differences between the DtxR homologs (Fig. 4.1). The Fe(II)-responsive

CdDtxR, MtIdeR and the Mn(II)-responsive SgScaR are comprised of three domains, while the

Mn(II)-responsive BsMntR lacks the third SH3-like domain of DtxR. The primary metalloregulatory site, as revealed by the structure of Fe(II)-responsive protein CdDtxR and

MtIdeR, involves four residues, M10, C102, E105 and H106 (using the CdDtxR numbering). In contrast, in the Mn(II)-responsive proteins M10 is replaced by D8 and C102 is replaced by E99

(using the BsMntR numbering). The D8 conserved in the Mn(II)-responsive protein has been shown to be essential for the selectivity of MntR for Mn(II) over Fe(II) and replacement of the

D8 with methionine (D8M) relaxes this specificity [176]. The C102 conserved in the Fe(II)- responsive protein affects the metal-binding affinity of CdDtxR and substitution of serine for

C102 (C102S) abolishes the repressor activity [171, 172].

The DtxR homolog in Thermococcus kodakarensis KOD1 is encoded by TK0107. It has

71% identity and 88% similarity with PF0851. Both lack the SH3-like domain of DtxR and contain the same set of conserved metal-binding residues, except for the two Asp residues in the

N-terminal sequence which are not present in PfDtxR (this will be discussed later in this chapter).

These include the D8 equivalent E9, which is essential for Mn(II)-selectivity in BsMntR, and the

C101 (TK0107 numbering), which is only conserved in Fe(II)-responsive CdDtxR and MtIdeR.

It has been reported that TK0107 regulates the transcription of a set of transporters in response to iron in Thermococcus kodakarensis KOD1 [151]. The high sequence similarities and the

61 conserved metal-binding residues suggest that PfDtxR may play a similar role in P. furiosus iron homeostasis.

4.2.1 His-tagged PfDtxR Protein Expression and Purification

To verify that PfDtxR was a sequence specific DNA-binding protein, the gene (PF0851) was overexpressed to give the his-tagged recombinant protein in E. coli for in vitro studies. A clone carrying the PfDtxR-expression vector was obtained from Dr. M. W. W. Adams,

University of Georgia. Native PfDtxR is predicted to contain 133 amino acids with a calculated molecular weight of 15687.51 Da. The recombinant PfDtxR contains Met-Ala-His6-Gly-Ser at the N-terminus and gives a molecular weight of 16725.57 Da. The BL21-CodonPlus (DE3)-

RIPL strain was utilized for efficient expression of PF0851 in E. coli host strain. Expression of the protein did not appear to be toxic to the cells and did not significantly slow culture growth.

The protein was expressed in abundance, predominantly in soluble form, which greatly simplified the purification procedure (Fig 4.3).

The his-tagged protein was purified from cell extract using a nickel-affinity column, and a gradient of imidazole was used to elute the protein from the column. The elution from nickel- affinity column were collected and pooled for buffer exchange using a centrifuge filter tube. An example gel showing protein purified in this manner can be seen in Figure 4.3. Once the protein was exchanged into a suitable salt-containing buffer, it was divided into small aliquots for storage and use; some of the aliquots were stored at -20 °C to be taken individually for short term use in in vitro assays, the rest were kept at -80 °C for long-term storage.

4.2.2 His-tagged PfDtxR-DNA Binding Affinity

In order to verify the binding ability of his-tagged DtxR to DNA, DNA probes containing putative promoter regions (~200-bp upstream and ~50-bp downstream of the ORF start) of its

62 own gene (PF0851) and the iron-regulated ferrous transporter operon feoAB (PF0858-57) were chosen as targets for the DtxR-DNA binding assay. No specific binding-shifts were observed in

EMSA when the protein to DNA ratio was increased from 8 to 32, indicating that DtxR does not recognize either of these promoters (Fig. 4.4). The effect of Fe on protein-DNA binding was also tested for DtxR with the feoAB and dtxR promoters. Ferrous ammonium sulfate was freshly prepared and added to the assay, which included 1mM DTT to keep the Fe2+ from being oxidized in the aerobically performed assay. No specific binding-complex was seen under the addition of

Fe2+; although the non-specific binding complex formed at protein to DNA ratio 32 was not seen in the presence of Fe (Fig. 4.4). Overall, the his-tagged DtxR failed to recognize the promoters of feoAB operon and dtxR even in the presence of Fe2+.

4.3 Reanalysis of the DtxR homolog in Pyrococcus furiosus

DtxR homologs were found in all Thermococcaceae species and they all exhibit a high- degree of sequence identity (Fig. 4.5). Consequently, further inspection of the sequence encoded by PF0851 indicated that there was a missing N-terminal extension in PF0851 compared to all other DtxR family members (Fig. 4.5). This extension contains two of the metal-binding residues that are otherwise conserved in the other orthologs. An analysis of the upstream sequence of

PF0851 gene revealed an alternative start codon, TTG, which is less common as start codon compared to ATG, but known to be functional in other genes [177, 178]. This is located 36~34- bp upstream of the ATG that is annotated as the start codon in the NCBI database. The translated protein sequence of the N-terminal extension is highly similar to those of the other

Thermococcaceae DtxR family members, including the conserved metal-binding sites Glu-8 and

Glu-11, suggesting that this may be part of the ORF. The Bacillus subtilis MntR protein adopts a common fold regardless of the bound metal ion. It consists of an N-terminal 70-residue DNA-

63 binding domain, containing a helix-turn-helix (HTH) motif, and a C-terminal dimerization domain, comprising residues 71-142 [179]. The PF0851 protein was expected to fold in the same way given the high sequence similarity (61%) (Fig. 4.6). The predicted secondary structure of

PF0851 (Fig. 4.6) suggests that the N-terminal extension is involved in the formation of the HTH motif. Mutations of the metal-binding residues have been reported to modify DtxR metal selectivity and responsiveness as metal-binding helps drive quaternary structural changes that mediate allosteric coupling between the metal and DNA binding sites resulting in dysfunctional regulators [107, 171, 172, 180, 181]. Therefore, the N-terminal extension discovered in PF0851 may play an important role for PF0851 to function as an iron-responsive regulator.

4.4 Validation of the DNA-binding affinity of PfDtxR

4.4.1Pf DtxR Protein Expression and Purification in E. coli

The gene encoding the PfDtxR protein with the N-terminal extension (named "full- length" PfDtxR) was overexpressed as a his-tagged recombinant protein in E. coli for in vitro studies. A PfDtxR-expression vector was constructed using a vector modified from pET24d, which allowed for the expression of PfDtxR with an N-terminal his6-tag to facilitate protein purification. Native PF0851 with the new start site contains 145 amino acids and a calculated molecular weight of 17181.16 Da. The recombinant PfDtxR with a his-tag contains Met-Ala-

His6-Gly-Ser at the N-terminus and the molecular weight is 18219.22 Da. The BL21-CodonPlus

(DE3)-RIPL strain was utilized for efficient expression of PfDtxR in E. coli host strain.

Expression of the protein did not appear to be toxic to the cells and did not significantly affect growth. Moreover, the protein was highly expressed as indicated by gel electrophoresis, predominantly in soluble form, which greatly simplified the purification procedure (Fig 4.7).

The his-tagged protein was purified from cell extract using a nickel-affinity column, and a gradient of imidazole was used to elute the protein from the column. The elution from nickel- affinity column involved several unwanted proteins and thus fractions with acceptable purity were collected and pooled for a gel filtration column for additional purification (Fig. 4.7).

Carbonic anhydrase (29 kDa) and albumin (66 KDa) were used as protein markers (Fig. 4.8).

The elution peak of his-tagged DtxR appeared between the peaks of carbonic anhydrase and albumin, indicating that his-tagged DtxR is a dimeric protein (~36 kDa). Fractions were collected and pooled for buffer exchange using centrifugal ultrafiltration. An example of the gel of the protein purified in this manner can be seen in Figure 4.7. Once the protein was exchanged into a suitable salt-containing buffer, glycerol was added to avoid freeze-induced structure damage and then divided into aliquots; part of the small aliquots were stored at -20 °C to be taken individually for short-term use in in vitro assays, and the rest were kept at -80 °C for long- term storage.

4.4.2 PfDtxR binds specifically to promoters of iron transporters

In order to verify the binding ability of his-tagged DtxR to DNA, DNA probes containing the putative promoter regions (~200-bp upstream and ~50-bp downstream of the ORF start, see

Table 2.1 for details) of iron-regulated genes were designed for protein-DNA binding assays.

Also included were genes regulated in response to iron according to DNA microarray data (Table

3.2): ftr1 (PF0723), feoAB (PF0858-PF0857), sufCDB (PF1287-PF1285) and aor (PF0346). The putative promoter region of PF2025 encoding the S0-induced protein SipA was also selected as it only responds to sulfide in the presence of iron [182]. The promoter of dtxR gene was also included to investigate auto-regulation.

EMSA was performed to determine whether DtxR binds specifically to any of the promoters of the iron-regulated genes. DtxR was found to bind to the promoters of ftr1, feoAB and sipA. For the promoters of ftr1 and feoAB, more than one distinct protein-DNA complex were observed in the range of protein-DNA ratios tested (Fig. 4.9). For example, three protein-

DNA complexes were observed at protein-DNA ratio as low as 4 for the ftr1 promoter, and all the free DNA along with the binding complexes were shifted at a protein/DNA ratio of 16; while the binding to feoAB promoter gave rise to only one complex that appeared at ratio 4. The second revealed with a smear of non-specific binding complexes at a ratio of 8 and even at 16 there was

DNA left unbound. One binding complex was observed for sipA starting at a ratio of only 4, the complex remained at a ratio of 8, but non-specific binding was evident as a smear at a ratio of 16.

The dtxR promoter had small binding-shift at a ratio of 4 which was seen as non-specific binding when protein-DNA ratio was increased to 8. No obvious binding was observed for aor, fur and suf probes in the range of protein-DNA ratios tested (Fig. 4.10).

To make sure the recognition of DtxR to the probe DNA was specific, a piece of DNA containing the putative promoter region (~-200-bp upstream and ~+50-bp downstream of the

ORF start, see Table 2.1 for details) of PF0849 gene, the transcription of which does not respond to iron, was used as a negative control in the EMSA with DtxR. As expected, under the same experimental condition, no binding complex was generated by DtxR with the promoter for this gene. At high protein-DNA ratios from 16 to 32, the probes were eventually shifted into a smear by the non-specific binding of DtxR (Fig. 4.9).

To further confirm the specificity of DtxR to the ftr1, feo and sipA promoters, heparin was used in EMSA as a non-specific binding competitor. Since the EMSA was based on nonradioactive method, unlabeled nucleic acid could not be applied as nonspecific competitor as

66 the competitor would interfere with the visualization of DNA target in the gel. Heparin is a sulfated polysaccharide and can be considered as a structural analog of the DNA phosphate backbone. It is able to bind most of the proteins recognizing DNA sequences non-specifically and was used as a non-specific binding competitor [183]. Base-specific contacts of a protein with

DNA have much higher affinity than pure electrostatic association with the phosphate-sugar backbone, therefore, a protein binding non-specifically to DNA solely through electrostatic associations should be competed off with heparin since it has a highly negative overall charge and resembles DNA. The fluorescent tag ethidium bromide that was used to visualize DNA in the gel does not intercalate with heparin, making it invisible in the assay. At heparin concentrations between 20 and 200 ng/μL, DtxR binding to the pf0849 probe was completely blocked. Given this finding, it is not surprising that at high protein/DNA mole ratios, DtxR associates non-specifically even with the promoters of ftr1 and feo, as evidenced by the shift to very low mobility complexes at protein/DNA mole ratios of 16 and 32. It is also worth noting that heparin at sufficiently high concentrations can also compete off sequence-specific DNA binding as shown by its effect on DtxR binding of ftr1 and feo promoter at a concentration of 2 mg/mL. The specific binding complex of sipA promoter was completely competed off at heparin concentration of 200 ng/μL, indicating a lower binding affinity of DtxR to sipA promoter, compared to other promoters.

The ability of MgCl2 to reduce protein-DNA binding [184] was also investigated in

EMSA. This confirmed that the binding complex observed for sipA promoter was the result of specific protein-DNA binding. Nonspecific protein-DNA binding is likely due to electrostatic attraction between positively charged amino acids of the protein and negatively charged phosphate oxygens of the DNA. The addition of the magnesium cation minimizes the

67 electrostatic attraction therefore reduces non-specific binding. Non-specific binding was significantly reduced to the pf0849 promoters when 10 mM MgCl2 was added to the assays, while the binding shifts from feoAB and sipA promoters were not affected (Fig. 4.11), indicating that binding to their promoters were the result of specific protein recognition.

The promoters of ftr1, feo and sipA were aligned to search for DNA motifs for which

DtxR had high affinity. The motif search was performed using MEME motif finding software

[185] in the forward and reverse strands of the three promoters and a graphical representation of the motif was generated using WebLogo [186]. A motif was revealed by the search (Fig. 4.12) containing a palindrome DNA sequence CCTAAn5TTAGG. This motif is located -25 to -10 bp upstream of the ftr1 ORF; in the feo promoter, the sequence is CCTAAn5TGAGG, with the second thymine replaced by guanine in the 3‟-terminal of the motif, ending up as an imperfect palindrome.

In order to further identify the position of the DNA motif within the promoters, DNA sequences were designed to contain three separate fragments, each of 80 to 90-bp, from the ftr1 promoter. The S1 fragment ranged from the -76 to +7-bp region of the ORF, covering the entire palindrome sequence CCTAAn5TTAGG, the S2 fragment contained the -177 to -81-bp upstream region, and the S3 fragment contained the -18 to +60-bp region including half of the palindrome

TTAGG (Fig. 4.12). As exhibited in the EMSA, DtxR evidently had a higher-affinity to the S1 fragment in the range of protein DNA ratio tested. A specific binding complex was observed from the S1 fragment at protein:DNA ratio of 4 and all the free DNA was shifted at a ratio of 8; while free DNA of the S2 and S3 fragments were not completely shifted even at a ratio of 32.

These results showed that the protein had the highest affinity for DNA sequence containing the

68 complete palindrome; although they do not prove that the palindrome is the bona fide protein- binding motif. Further work needs to be carried out to determine the true DtxR operator.

4.4.3 PfDtxR DNA-binding responds to iron

The effect of metals on DtxR-DNA interaction was also determined in this study using

EMSA. Since the DtxR homolog in Thermococcus kodakarensis TK0107 was reported to regulate transcription in response to iron in vivo [151], Fe2+ was first added in the assays for possible effects on binding (Fig. 4.13). As a result, the specific binding shifts of feoAB promoter were no longer observed in the assays containing 500 μM Fe2+ at protein:DNA ratios below 32, comparing alongside with assays of pf0849 promoter in which the non-specific binding at high protein DNA ratios was also inhibited. It was concluded that binding of DtxR to cognate DNA was abolished in the presence of iron.

Investigation of the effect of Mn2+, Co2+ and Zn2+ on DtxR DNA binding was surveyed using assays containing the corresponding cation with feoAB promoter (Fig. 4.13). It was obvious that with the same concentration added, binding of DtxR to feoAB promoter was retained in the presence of all metal cations except for Fe2+. Therefore, it was concluded that the

DNA-binding ability of DtxR can be abolished exclusively by excess amount of Fe2+ in vitro.

If iron abolished the binding of protein to the promoters of iron transporters in vivo, DtxR would need to be a transcriptional activator for those genes so that the transporters are up- regulated at low iron concentration in P. furiosus. However, the concentration of Fe2+ in EMSA is well above the physiological range in our in vitro studies, so that the relation to in vivo behavior cannot yet be established. There may also be a concern that the inhibition of protein-

DNA binding could be the effect of protein damage induced by iron [138]. To learn about the

69 cellular role that DtxR may play in vivo and the way it responds to iron, the dtxR deletion mutant was constructed and characterized.

4.5 Characterization of the dtxR deletion mutant

4.5.1 Construction of the dtxR deletion mutant

A marker-replaced deletion of dtxR (DTXR) was constructed in the COM1 background strain (ΔpyrF) based on published methods [113]. The COM1 strain lacks the pyrF gene which encoding the uracil biosynthetic enzyme orotidine-5‟-monophosphate (OMP) decarboxylase, thus allowing uracil prototrophic selection. A linear DNA fragment was constructed by overlapping PCR containing the pyrF (PF1114) gene under the control of the P. furiosus glutamate dehydrogenase gdh (PF1602) promoter (PgdhpyrF cassette) flanked with ~1kb flanking region of the dtxR gene for homologous recombination with the downstream flanking region containing the last 30 bp of dtxR at the C-terminal to keep the intact PF0852 gene, which overlaps with dtxR (Fig. 4.14). The linear DNA was then transformed into mid-exponential phase

COM1, which is naturally competent for DNA uptake [113]. The dtxR gene in COM1 is replaced by the PgdhpyrF cassette on the linear DNA fragment by homologous recombination. The selection for transformed colony was performed on defined media without uracil since the marker-replaced ΔdtxR strain is able to grow in the absence of uracil. Experiments using COM1 alone plated on media containing or lacking uracil were performed as positive and negative controls, respectively. Approximately 105 transformants were obtained per μg DNA per 108 cells, and hundreds of transformants were obtained by simply mixing the linear DNA fragment containing a PgdhpyrF cassette with a small volume of culture (2 ng DNA per μL culture) under aerobic conditions and plating this mixture onto defined medium without uracil for selection of transformants under anaerobic growth conditions at 90 to 98 ºC. No colonies were observed in

70 the absence of added DNA on the negative control plates, even when plating over 100-fold more cells than those used for transformation. The pH of the solid media was key to the successful transformation ability of COM1 strain. For example, when the cellulose-based defined media was made at pH 6.8, no transformants were observed after incubation for 60 to 65 h; but when the pH of the same media was adjusted to pH 6.7, hundreds of transformants were observed.

Colonies found on the positive control plates at pH 6.8 were much smaller than those on the plates at pH 6.7, indicating the tiny shift of pH could have an enormous impact on the ability of cell growth and natural competence of COM1. Colonies were picked and purified in defined media lacking uracil, and the resulting strain was confirmed using PCR and sequence analyses for gene replacement. In addition, the qPCR product of the deleted gene was not detected (Fig.

4.15). Finally, the ΔdtxR strain was grown in complex media and harvested in the mid- exponential phase to make glycerol stocks. It was designated as strain DTXR.

4.5.2 Growth phenotype of dtxR deletion mutant

To test the effect of iron limitation on the DTXR mutant, growth was analyzed in iron- rich medium containing 10 μM Fe(II) and iron-limited medium with no iron added (the actual Fe concentration is < 0.8 μM as measured by ICP-MS). Inocula used in this experiment were 2%

(vol/vol) of the active cultures that had been passed twice in medium with no iron added to limit carryover of excess iron. The resulting growth curves showed reduced growth rate in iron-limited medium, but the curves using both iron levels were similar for the COM1C2 (pyrF::pyrF) and

DTXR strains (Fig. 4.16). The Fe2+-specific chelator BPS was added to the iron-limited medium to create a more stringent environment for P. furiosus to survive. The growth was similar to that in iron-limited medium when 10 μM BPS was added but cells showed poor growth when the BPS concentration was increased to 30 μM. In either case, no phenotype was

71 observed for DTXR under these more stringent growth conditions. It was concluded that

DTXR did not exhibit any growth advantage or disadvantage compared to COM1C2 when iron concentration varied.

4.5.3 Microarray analysis of transcriptional response to iron in DTXR

Iron as an essential element for most life forms. It is a cofactor for many enzymes and is involved in vital metabolic processes. The P. furiosus genome encodes a high number of iron- containing proteins involved in energy metabolism, H2 production, oxidative stress detoxification,

DNA replication and so on, which emphasizes the important role that this element plays. The requirement of iron remains uncharacterized for P. furiosus; in anaerobic environments, enough iron is expected to be present in the form of Fe(II) form that its accessibility should not be limiting. In addition, Fe(II)-mediated formation of reactive oxygen species via Fenton chemistry is not present in the absence of oxygen. Thus, elaborate iron regulation would seem to be less critical in anaerobes. However, the genome of Desulfovibrio vulgaris, a sulfate-reducing anaerobic bacteria, contains three fur paralogs, fur (DVU0942), perR (DVU3095) and zur

(DVU1340). A diverse transcriptional response was observed for the fur deletion mutant compared to the wild-type D. vulgaris. The Fur regulon of this anaerobe includes ferrous iron transporter genes feoAB, flavodoxin fld and P-type and ABC ATPases and biopolymer transporter tonB [187, 188]. A homolog of the iron-responsive transcription regulator DtxR was found in the P. furiosus genome and in vitro DNA-binding studies presented earlier showed that

DtxR has an iron-responsive binding ability to promoters of the iron transporter genes indicating a potential role in iron regulation. In order to characterize growth of P. furiosus in response to iron deficiency, as well as identifying the role of DtxR in transcription regulation, we investigated the global regulatory gene response in DTXR using microarray analysis.

As observed from previous growth studies (Fig. 4.16), even though growth of P. furiosus was much better in iron-rich medium, the growth rates in exponential phase in iron-rich and iron- limited medium were similar, indicating that sufficient iron was available in the latter at that stage. As growth continued, the “iron pool” started to drain in the iron-limited medium, thus growth slowed down and entered stationary phase earlier; while in the iron-rich medium the excess amount of iron supports the growth until other nutrients in the medium became limiting, and a longer exponential phase is exhibited.

Cultures of COM1C2 and DTXR were grown in media containing 10 μM Fe(II) or with no iron added, and harvested in late exponential phase (Fig. 4.16). It was assumed that the transcription response would be the most dramatic between the two conditions in cells entering stationary phase. Samples were taken at similar optical densities for both COM1C2 and DTXR, with differential gene expression calculated as log2 ratios using the following formula: log2 (iron- limited) – log2 (iron-rich).

A total of 2,193 ORFs are annotated in the genome sequence of P. furiosus.

Approximately 30% of these are designated (conserved) hypothetical and show no similarity to characterized ORFs in other genomes [189]. A DNA microarray had been previously constructed using PCR products of each of the 2,193 ORFs. The arrays were used to assess differential gene expression in our strains grown in the presence or absence of iron and were provided by Dr.

Gerrit Schut of University of Georgia [190].

Results were obtained from RNA samples that were prepared from two different P. furiosus mutant strains, the expression profiles were compared in two cultures of the same strain grown independently in iron-rich and iron-limited media. The results of a control experiment using RNA samples prepared from two different cultures of COM1C2 grown under identical

73 conditions are shown in Figure 4.17.A. The fluorescence signal intensities vary over more than a

103 range. ORFs with intensities less than 500 arbitrary units are considered not to be expressed at a significant level. As expected, the low-intensity signals show a high standard deviation, due to background fluorescence, while the more-highly expressed ORFs lie close to the diagonal. As indicated by lines showing a twofold change, most ORFs fall within this range. For COM1C2 and DTXR, approximately 70% (1,523 of 2,193) and 77% (1,700 of 2,193) of the ORFs respectively are significantly expressed (> 500 units) under one or both growth condition.

The general transcription profiles of the genes in COM1C2 were similar under high- and low-iron conditions, showing 44 ORFs regulated (30 up- and 14 down-regulated) by more than twofold in response to iron (Tables 4.1 and 4.2). The transcription of putative iron transporters ftr1 (PF0723) and feoAB (PF0858-PF0857) increased in the iron-limited condition. Interestingly, a homolog of a phosphate transport system (PF1020-PF1021) was also found up-regulated in response to iron limitation, suggesting that this system might be implicated in iron transport. The metabolism of amino acids was also affected by iron. Operons involved in branched amino acid biosynthesis (PF0934-PF0942) and histidine biosynthesis (PF1657-PF1666) were both up- regulated when iron is limiting. Meanwhile, three ABC-type transporters annotated as peptide transporter (PF0191-PF0196), cobalt transporter (PF0528-PF0531) and maltose transporter

(PF1936-PF1938) were down-regulated. The iron-containing enzyme aldehyde:ferredoxin oxidoreductase (AOR), a known tungsten-containing protein that oxidizes aldehyde [75] was down-regulated when iron was limiting. Other ORFs appear to be related to oxidative stress. For example, PF1033 is homologous to peroxiredoxin in P. horikoshii (PH1217), which catalyzes the reduction of cumene hydroperoxide and hydrogen peroxide using dithiothreitol as an electron donor [191]. The expression of PF1033 and PH1217 were both up-regulated by oxidative stress

[94, 191], indicating an important role in peroxide-scavenging system. As the iron level in the system decreases, the chances of Fenton reaction [192] and the generation of highly reactive oxygen species are lowered, hence it is reasonable for P. furiosus to tune down its scavenging system when cells are not threatened by oxidative stress.

In the DTXR strain, when iron was made limiting, the expression of 15 genes increased and 8 were decreased more than twofold, and these are listed in Table 4.1 and 4.2. A comparison between the DTXR transcription profile and that in COM1C2 (Figure 4.18) showed that all the

23 genes regulated in DTXR could be found regulated in COM1C2 illustrating that the transcription of these genes respond to iron regardless of the presence of DtxR. Therefore, the genes differentially regulated in COM1C2 and DTXR are candidates for the DtxR regulon. A total of 9 operons met this criterion and showed changes in regulation patterns in the two strains

(Table 4.1 and 4.2, ORFs shaded in grey; Fig. 4.18). These included the putative iron transporters ftr1 (PF0723) and feoAB (PF0858-PF0857), the putative phosphate transport system

(PF1020-PF1021) and cobalt transporter (PF0528-PF0531), and the iron-containing enzyme

AOR (PF0346).

Among these genes, PfDtxR has been shown to bind specifically to the promoters of ftr1 and feoAB, which agrees with the result of the microarray analysis. The fluorescence signals of two iron transporters from microarray data indicate a repressed transcription in COM1C2 in iron- rich medium. For example, the fluorescence signal of ftr1 in COM1C2 was 740 units compared to the signal of 1440 units in DTXR grown under iron-rich conditions. These microarray results were confirmed using qPCR.

4.5.4 Quantitative PCR confirmation of PfDtxR as a repressor

To confirm the regulatory role of DtxR in P. furiosus, qPCR was performed using the same cDNA derived from COM1C2 and DTXR in iron-rich and iron-limited conditions for microarray analysis. The transcription of ftr1 (PF0723) and feoAB (PF0857-58) was compared between the COM1C2 and DTXR strains under the same growth condition to determine the role of DtxR as a transcription regulator in P. furiosus. The levels of transcription for the two genes were similar in DTXR and COM1C2 under iron-limited conditions. In contrast, under iron-rich conditions, the levels of gene transcription in DTXR were higher than those in

COM1C2, indicating DtxR is an iron-dependent transcription repressor for the iron transporters in P. furiosus.

The result of qPCR experiments confirmed the fact that PfDtxR specifically binds to the promoters of ftr1 and feoAB and based on the pattern of regulation in the deletion mutant, it was concluded that DtxR binds to DNA in the presence of iron and plays the role as a transcription repressor for the putative iron transporters in P. furiosus.

Table 4.1 ORFs whose expression is > 2-fold up-regulated in Fe-limited condition and their potential operon arrangement

COMC2 DTXR a b COMC2 DTXR ORF Description c Fold c Fold log2 ± SD d log ± SD change 2 changed PF0685 conserved hypothetical protein 1.2 ± 0.1 2.2 1.2 ± 0.4 2.3 PF0686 hypothetical protein 0.6 ± 0.2 1.5 0.9 ± 0.4 1.9 PF0687 Transcription factor B-2 1.0 ± 0.2 2.0 1.1 ± 0.4 2.2

PF0692 Prismane 1.1 ± 0.1 2.1 1.7 ± 0.3 3.2

PF0694 flavoprotein 0.4 ± 0.4 1.3 0.9 ± 0.8 1.8 PF0695 hypothetical protein 1.0 ± 0.2 2.0 1.8 ± 0.6 3.5 PF0696 carbohydrate-binding protein 0.5 ± 0.2 1.4 0.6 ± 0.9 1.5

PF0723 FTR1 iron permease 1.6 ± 0.2 3.1 0.8 ± 0.3 1.7

PF0728 hypothetical protein 1.1 ± 0.4 2.1 1.2 ± 0.4 2.3 PF0729 multi domain protein 0.3 ± 0.7 1.2 1.1 ± 0.5 2.1

PF0784 conserved hypothetical protein 1.2 ± 0.3 2.4 -0.1 ± 0.3 1.1

Putative ferrous transporter PF0857 ferrous iron transport protein B 1.0 ± 0.0 2.0 -0.1 ± 0.1 1.0 PF0858 conserved hypothetical 0.3 ± 0.3 1.2 -0.3 ± 0.3 1.2

Branched amino acids biosynthesis PF0934 hypothetical 1.8 ± 0.3 3.6 1.0 ± 0.5 1.9 PF0935 acetolactate synthase 2.2 ± 0.6 4.6 1.7 ± 0.4 3.3 PF0936 ketol-acid reductoisomerase 2.5 ± 0.6 5.5 1.8 ± 0.4 3.4 PF0937 2-isopropylmalate synthase 2.3 ± 0.5 4.9 1.7 ± 0.3 3.3 PF0938 3-isopropyl malate dehydratase II 2.2 ± 0.5 4.7 1.7 ± 0.4 3.3 putative 3-isopropylmalate 1.9 ± 0.5 3.8 1.4 ± 0.3 2.7 PF0939 dehydratase PF0940 probable isocitrate dehydrogenase 2.4 ± 0.5 5.3 1.5 ± 0.4 2.8 PF0941 alpha-isopropylmalate synthase 2.5 ± 0.5 5.5 1.7 ± 0.2 3.3 PF0942 dihydroxy-acid dehydratase 2.4 ± 0.5 5.2 1.5 ± 0.2 2.8 Continued on following page

Table 4.1 – Continued

COMC2 DTXR a b COMC2 DTXR ORF Description c Fold c Fold log2 ± SD d log ± SD change 2 changed Putative phosphate transporter PF1020 probable phosphate transport protein 1.6 ± 0.4 3.1 0.0 ± 0.5 1.0 PF1021 conserved hypothetical protein 1.4 ± 0.3 2.7 0.1 ± 0.3 1.1

PF1085 acetyl -CoA synthetase Q3 alpha 1.8 ± 0.2 3.5 0.6 ± 0.1 1.5

PF1085.1 hypothetical 1.8 ± 0.4 3.4 0.1 ± 0.3 1.1

PF1404 proteasome, subunit beta 1.0 ± 0.4 2.0 0.8 ± 0.3 1.7 PF1405 polyadenylation specifity factor 1.4 ± 0.5 2.7 1.0 ± 0.3 2.1 PF1406 probable threonine synthase 1.0 ± 0.5 2.0 0.9 ± 0.3 1.9

PF1616 myo -inositol-1-phosphate synthetase 1.2 ± 0.2 2.3 0.8 ± 0.3 1.7

Histidine biosynthesis PF1657 histidyl-tRNA synthetase 1.1 ± 0.6 2.2 0.7 ± 0.3 1.6 PF1658 ATP phosphoribosyltransferase 0.9 ± 0.6 1.8 0.5 ± 0.3 1.4 PF1659 histidinol dehydrogenase 1.0 ± 0.5 2.0 0.7 ± 0.2 1.6 PF1660 imidazoleglycerolphosphatedehydratase 0.8 ± 0.6 1.8 0.7 ± 0.3 1.6 PF1661 glutamine amidotransferase 0.8 ± 0.5 1.7 0.5 ± 0.3 1.4 PF1662 HisA 1.1 ± 0.4 2.1 0.5 ± 0.3 1.4 PF1663 imidazoleglycerol-phosphate synthase 0.7 ± 0.4 1.7 0.4 ± 0.2 1.4 PF1664 phosphoribosyl-AMP cyclohydrolase 1.3 ± 0.4 2.5 0.6 ± 0.2 1.5 PF1665 histidinol-phosphate aminotransferase 1.0 ± 0.4 2.0 0.4 ± 0.2 1.3 PF1666 conserved hypothetical protein 1.0 ± 0.3 2.0 0.4 ± 0.2 1.3

PF1898 conserved hypothetical protein 1.2 ± 0.3 2.3 0.8 ± 0.3 1.7

PF1951 Archaeal asparagine synthetase A 1.1 ± 0.2 2.1 -0.1 ± 0.2 1.1 a The ORF description is derived from the annotation in NCBI and TIGR databases. Potential operons are indicated by bold entries within a group where the intergenic distances are less than 30 nt. b Description derived from NCBI database. c The intensity ratio (-Fe/+Fe ) is expressed as a log2 value so that the standard deviation can be given. For comparison between ORFs, the apparent change in the expression level is also indicated. ORFs are listed that are more than twofold regulated or that are potentially part of an operon with twofold-regulated ORFs but which themselves are regulated by at least twofold. d Calculated from the average log2 intensity ratio. Genes significantly differentially regulated in COM1 and DTXR were shaded in grey.

Table 4.2 ORFs whose expression is > 2-fold down-regulated in Fe-limit condition and their potential operon arrangement

COMC2 DTXR a b COMC2 DTXR ORF Description c Fold c Fold log2 ± SD d log ± SD Change 2 Changed Putative peptide transporter PF0191 peptide transporter -0.8 ± 0.5 1/1.8 -0.6 ± 0.4 1/1.5 PF0192 oligopeptide transport -1.3 ± 0.5 1/2.5 -1.1 ± 0.2 1/2.2 PF0193 putative ABC transport -1.1 ± 0.6 1/2.1 -1.0 ± 0.2 1/2.0 PF0194 dipeptide ABC transporter -1.2 ± 0.6 1/2.2 -1.1 ± 0.4 1/2.1 PF0195 conserved hypothetical protein -0.7 ± 0.4 1/1.7 -0.8 ± 0.3 1/1.7 PF0196 glucose-6-phosphate isomerase -0.4 ± 0.1 1/1.3 -0.3 ± 0.3 1/1.2

PF0341 conserved hypothetical protein -1.0 ± 0.5 1/2.0 -0.7 ± 0.9 1/1.6

PF0346 aldehyde ferredoxin oxidoreductase -1.6 ± 0.4 1/3.0 -0.9 ± 0.3 1/1.8

Putative cobalt transporter PF0528 cobalt transport protein -0.8 ± 0.3 1/1.7 0.2 ± 0.2 1.1 PF0529 conserved hypothetical protein -0.5 ± 0.3 1/1.4 0.2 ± 0.2 1.1 PF0530 conserved hypothetical protein -0.8 ± 0.3 1/1.7 0.0 ± 0.3 1.0 PF0531 cobalamin biosynthesis protein -1.0 ± 0.5 1/2.0 0.0 ± 0.2 1.0

PF0678 conserved hypothetical -1.0 ± 0.4 1/2.0 -0.5 ± 0.2 1/1.4

PF1032 cys rich ORF -1.2 ± 0.6 1/2.3 -0.7 ± 0.3 1/1.6

PF1033 peroxiredoxin -2.1 ± 0.4 1/4.3 -2.0 ± 0.4 1/4.0

PF1242 molybdopterin oxidoreductase -1.2 ± 0.1 1/2.2 -1.0 ± 0.3 1/2.0 PF1243 conserved hypothetical -1.0 ± 0.3 1/2.0 -0.7 ± 0.4 1/1.7 PF1244 Hypothetical -1.0 ± 0.2 1/1.9 -0.8 ± 0.4 1/1.8

PF1890 conserved hypothetical protein -1.1 ± 0.1 1/2.1 -0.4 ± 0.3 1/1.4

Putative sugar transporter PF1936 putative sugar transport protein -1.2 ± 0.5 1/2.3 -1.2 ± 0.5 1/2.2 PF1937 putative sugar transport protein -1.3 ± 0.4 1/2.4 -1.1 ± 0.5 1/2.2 PF1938 maltodextrin binding protein -1.2 ± 0.6 1/2.3 -1.1 ± 0.4 1/2.1 a The ORF description is derived either from the annotation in NCBI or TIGR databases. Potential operons are indicated by bold entries within a group where the intergenic distances are less than 30 nt. b Description derived from NCBI database.

79 c The intensity ratio (-Fe/+Fe ) is expressed as a log2 value so that the standard deviation can be given. For comparison between ORFs, the apparent change in the expression level is also indicated. ORFs are listed that are more than twofold regulated or that are potentially part of an operon with twofold-regulated ORFs but which themselves are regulated by at least twofold. d Calculated from the average log2 intensity ratio. Genes significantly differentially regulated in COM1 and DTXR were shaded in grey.

Table 4.3 Quantitative PCR analysis of the regulation of iron transporters in DTXR and COMC2

DTXR/COMC2c ORFa Descriptionb iron-rich iron-limited PF0723 FTR1 iron permease 2.3 0.9 PF0858 Ferrous iron transporter A 2.0 0.9 a The ORF description is derived either from the annotation in NCBI or TIGR databases. b Description derived from NCBI database. c Quantitative PCR (standard PF0971) given as fold regulation of selected ORFs comparing DTXR to COM1C2 strain.

Figure 4.1 Structurally characterized DtxR-family proteins. A. Multiple sequence alignment of DtxR-family proteins from different species using CLUSTALW2. CdDtxR, C. diphtheria

DtxR; MtIdeR, M. tuberculosis IdeR; SgScaR, S. gordonii ScaR; BsMntR, B. subtilis MntR.

Metal-binding sites determined based on structural studies on these proteins are color coded. The same metal-binding sites containing well-conserved residues were found in CdDtxR and MtIdeR; different sites are colored in red and cyan, respectively. Different sets of metal-binding sites were found in SgScaR and BsMntR. The residues involved in metal-binding in SgScaR are colored in green; and those in BsMntR are colored in purple. B. Structures of (I) IdeR bound to Co2+ (PDB

ID 1F57 [193]); (II) MntR bound to Mn2+ (PDB ID 2F5D [179]) and (III) ScaR bound to Cd2+

(PDB ID 3HRT [169]). Side chains for metal-binding residues are shown for one subunit of each dimer as are bound metal ions (in purple). The N-terminal, DNA binding domains are colored yellow, the dimerization domains are in blue, and in IdeR and ScaR the C-terminal domains are colored green [169]. C. Phylogenetic distribution of DtxR family proteins. DtxR is conserved in phylogenetically diverse prokaryotes spanning both the Bacterial and Archaeal domains of life.

The phylogenetic tree was generated at the STRING site (http://string.embl.de/) using orthology information from the COG database.

Figure 4.2 PfDtxR (PF0851) protein sequence analysis. A. Conserved domain search results for DtxR protein sequence from online tools available at NCBI. PfDtxR sequence is represented by a line segment colored in grey and blue with matching conserved domains indicated below.

Conserved domain descriptions and sequence alignments are shown. For the sequence alignments, identical residues are colored red and similar residues are colored blue. B. Multiple sequence alignment of PfDtxR with CdDtxR, MtIdeR, SgScaR, BsMntR and TkDtxR

(TK0107p). Metal-binding sites are color coded the same way as in Figure 4.1. Conserved putative metal-binding residues are shaded in grey for PfDtxR and TkDtxR.

Figure 4.3 His-tagged PfDtxR protein expression and purification. A. SDS-PAGE of PfFur expressed in E. coli BL21. Lane 1: before induction, lane 2: after induction (4 h) with IPTG. B.

SDS-PAGE gel of fractions collected from the nickel-affinity chromatography step. The fractions were pooled for buffer-exchange into 20 mM sodium phosphate, 150 mM NaCl, pH 7.4.

A B

Figure 4.4 PfDtxR DNA-binding with dtxR and feoAB promoters. 100 nM promoter of (A) dtxR or (B) feoAB was incubated with PfDtxR at corresponding Protein/Nucleic acid (P/N) ratios shown above each lane.

A B

Figure 4.5 Analysis of the genomic region upstream of PF0851 reveals a new start site. A.

The PF0851 ORF sequence led by the new start site “TTG” and an N-terminal extension not predicted by the NCBI and TIGR databases, shaded in grey. B. DtxR homologs were found in all

Thermococcaceae species, sequence identities were calculated using BLAST. C. Protein sequence alignment of DtxR homologs from Thermococcaceae species using CLUSTALW2. For

PF0851, the N-terminal extension sequence are shaded in grey, the conserved metal-binding residues are colored in yellow.

A TTGCCATCAAAGAGGGAGGAAGAATATCTGGAAACCATGTATATCCTTCAAAAGAACAAGGGTGTAATAA GGGTAAAGGACATAGCCAAGATGATGAGGGTAAAACCCCCGACCGTTGTTGAGGCATTAAAAAAGCTTAG GGATAAGGGGTTTGTAAAATATGAGGAGCACGAGCACATTCTTCTCACTGAAAAAGGTCTGGAAGTTGCA AAGAAAACATACTCGAAGCATCAGCTTTTGACGGAGTTTTTCATAAACATTTTAGGCATTCCACCCGAAA TAGCTGAGAGAGATGCGTGCCAATTTGAGCACTACGTTAGTGAGGTTACAGTACATAGGATTAGAGAATT CATAAGCTATATTCAGCAAGAATGTCCCTATGCGTTAAAGCAGTTCTTAAAAAAAGTTAGAGAAAAGGAT CAAGCTGTGGCTAAGTAG

Figure 4.6 Reanalysis of PF0851 protein sequence. A. Protein sequence alignment of PF0851 with BsMntR (B. subtilis MntR) using the NCBI online sequence alignment tool. Conserved residues are indicated in the row between the two protein sequences and similar residues are indicated using + sign in the middle row. B. PF0851p secondary structure predicted using

PSIPRED [41]. The N-terminal extension is predicted to take part in forming the first helix in the

DNA-binding domain.

Figure 4.7 His-tagged PfDtxR protein expression and purification. A. SDS-PAGE of PfDtxR expressed in E. coli lane 1: before induction, lane 2: after induction (1 h) with IPTG. B. SDS-

PAGE of eluted fractions from the automated chromatographic purification of his6-PfDtxR from soluble cell extract using a gel filtration column with Washing Buffer (50 mM sodium phosphate pH 7.3, 0.15 M NaCl). Fractions B1-B6 were collected for further analysis (green). C. SDS-

PAGE gel of fractions collected from elution step of gel filtration chromatography. The fractions were pooled for buffer-exchange into 50 mM sodium phosphate, 150 mM NaCl, pH 7.4.

A B

Figure 4.8 Gel filtration analysis of PfDtxR. A. Gel filtration with collection of eluted fractions of PfDtxR from Ni-affinity column. Protein markers (B) albumin (66 KDa) and (C) carbonic anhydrase (29 kDa). Wavelength of the UV light was set at 280nm to detect eluted protein fractions (blue); flow detector (yellow) was used to monitor the consistency of flow rate

(1.5 ml/min). Fraction numbers were indicated above the X-axis (red); fractions B1-B6 were collected for buffer exchange and storage. UV2 254nm and UV3 215nm in the diagrams were parameters preset by ÄKTA purifier but not considered in this experiment.

B C

Figure 4.9 DNA-binding affinity of PfDtxR to promoter DNA. PfDtxR-DNA binding affinity was determined using promoters (-200 bp upstream to +50 bp downstream regions of the ORF start site) of (A) ftr1, (B) pf0849 (negative control), (C) feoAB and (D) sipA. In lanes 1 to 10, 100 nM promoters were incubated with PfDtxR at corresponding Protein/Nucleic (P/N) acid ratios listed above each lane (grey bar); in lanes 6 to 10, 1μL heparin was added from 0.2~2000 ng/μL

(black bar) to assays with P/N ratio at 32.

A B

C D

100

Figure 4.10 DNA-binding affinity of PfDtxR to promoter DNA. PfDtxR-DNA binding affinity was determined using promoters (-200 bp upstream to +50 bp downstream regions of the

ORF start site) of (A) dtxR, (B) aor, fur and suf as indicated above the diagrams. 100nM promoters were incubated with PfDtxR at corresponding Protein/Nucleic (P/N) acid ratios listed above each lane (grey bar).

101

A B

102

Figure 4.11 DNA-binding affinity of PfDtxR to promoter DNA. PfDtxR-DNA binding affinity was determined using promoters (-200 bp upstream to +50 bp downstream regions of the

ORF start site) of feo (lane 1-4 and 13-16), sipA (lane 5-8 and 17-20) and pf0849 (lane 9-12 and

21-24) as indicated above the diagrams. 100nM promoters were incubated with PfDtxR at corresponding Protein/Nucleic (P/N) acid ratios listed above each lane (grey bar). In lane 13-24,

1μl magnesium chloride was added to final concentration of 10mM in each assay.

103

104

Figure 4.12 Motif prediction and DNA-binding affinity of PfDtxR to DNA motif. A. The result of a motif search within promoters of ftr1, feo and sipA using MEME [185]. The palindromic sequence CCTAAn5TTAGG was outlined in the WebLogo diagram [186] and the promoter sequences below (in blue). Mismatched bases were underlined in the sequences.

Translation start sites are in bold. B. Promoter of ftr1 was divided into three fragments S1, S2 and S3. The S1 fragment ranged from -76 to +7-bp region of the ORF start site, containing the entire palindrome sequence CCTAAn5TTAGG; S2 fragment was -177 to -81-bp upstream region of the ORF and S3 fragment contained the -18 to +60-bp region of the ORF including half of the palindrome TTAGG. C. PfDtxR-DNA binding affinity was determined with fragment S1~S3. As indicated in the diagrams, 200 nM DNA were incubated with PfDtxR at corresponding

Protein/Nucleic (P/N) acid ratios listed above each lane (grey bar).

105

106

Figure 4.13 PfDtxR DNA-binding affinity in response to metals. PfDtxR-DNA binding affinity was determined using promoters (-200 bp upstream to +50 bp downstream regions of the

ORF start site) of (A) feo and (B) pf0849 as indicated above the diagrams. 100 nM promoters were incubated with PfDtxR at corresponding Protein/Nucleic (P/N) acid ratios listed above each lane (grey bar). In lane 6-10 in both diagrams, 1μL ferrous ammonium sulfate was added to a final concentration of 500 μM in each assay. C. Effect of metals on PfDtxR DNA-binding affinity. 100 nM feo promoter was incubated with PfDtxR at corresponding Protein/Nucleic (P/N) acid ratios listed above each lane (grey bar). Effects of no metal (lane 1, 2); Mn2+ (lane 3, 4);

Co2+ (lane 5, 6); Fe2+ (lane 7, 8) and Zn2+ (lane 9, 10) was determined. Metals were added to a final concentration of 500 μM in each assay.

107

A B

108

Figure 4.14 Construction of DTXR mutant strain. The linear DNA fragment used for transformation was constructed by overlapping PCR containing the pyrF (PF1114) gene under the control of the P. furiosus glutamate dehydrogenase gdh (PF1602) promoter (PgdhpyrF cassette) flanked with ~1kb flanking region of the dtxR gene. A. Genome region containing the dtxR (PF0851) gene. B. The linear DNA fragment containing 1kb flanking region of dtxR gene between the PgdhpyrF cassette. 30-bp in the C-terminal of dtxR gene was left in the genome

(PF0851* in diagram) to keep the T1 terminator sequence of PF0852 (PF0852 T) complete for transcription. Green arrow, ORFs predicted by NCBI; blue arrow, the “corrected” PF0851 ORF with the N-terminal extension. ORF numbers are indicated above the genome map in bold and primers (ZY001/007 and ZY005/006) designed to amplify the 1kb flanking region of dtxR gene were indicated by navy arrows. The gdh promoter is shown as a box in stripe and pyrF gene was illustrated as purple box in the genome map. Primers ZY011/012 were used for deletion confirmation by PCR.

109

PvuII (3400)

ZY007 ZY006 NcoI (3620) BglII (1713) NheI (3858) XbaI (1542) BmtI (3862) XhoI (1522) EcoRV (3948) AvaI (1522) AvaI (4081) PF0849 PF0850 PF0851 PF0853 XbaI (335) ZY001 ZY005 PvuII (4142) NdeI (317) NcoI (976) PstI (4147) PF0852 T

AE009950 4402 bp

PF0851 PF0852

XbaI (1171)

BglII (907) XbaI (736) AvaI (716) XhoI (716) PF 0851* PF 0852

PF 0850

PvuII (3108) p yrF Pg d h T1 terminator NcoI (3328) NcoI (170) ZY011 ZY001 ZY005

AE009950 3550 bp

ZY007 ZY006 ZY012

110

Figure 4.15 Verification of the dtxR deletion strain. A. Colonies from the second crossover event were picked and genomic DNA was isolated. Primers (ZY011/012) were designed to amplify the region containing the 1kb dtxR flanking region. B. Agarose DNA gel was performed to confirm their length, as the PCR product from the marker-replaced dtxR deletion strain should be longer than the original sequence (lane 5, 6, 7).

111

112

Figure 4.16 Growth of DTXR in response to iron. Growth curves of P. furiosus COM1C2

(diamond) and deletion strain DTXR (triangle) under iron-rich conditions (10 μM Fe(II), blue), iron-limited conditions (no iron added, red) and iron-depleted conditions (10 μM BPS, green; 30

μM BPS, purple). Black arrow indicates culture harvest point for RNA isolation for DNA microarray analysis. Inocula for the growth were passed three times in YEM-Fe to remove the residual iron associated with cells. All glassware was washed with 2% nitric acid overnight.

Optical density of the culture was measured for COM1 and DTXR at 660 nm.

113

Growth of DTXR in Response to Iron 0.30 0.25

0.20

0.15 O.D 660 660 O.D 0.10 0.05 0.00 0 2 4 6 8 10 12 14 Time (hour)

114

Figure 4.17 Relative fluorescence intensities of DNA microarrays. A. Comparison of cDNAs derived from two independent cultures of COM1 cells grown in an iron-rich medium. B.

Comparison of cDNAs derived from two independent cultures of COM1 cells grown in iron-rich and iron-limited media. C. Comparison of cDNAs derived from two independent cultures of

DTXR cells grown in iron-rich and iron-limited media. The upper and lower diagonal lines indicate 2-fold changes in the signal intensities. See Section 4.5.3 for details.

115

1000000

100000

10000

1000

100 100 1000 10000 100000

100000

10000

1000

100 100 1000 10000 100000

116

1000000

100000

10000

1000

100 100 1000 10000 100000

117

Figure 4.18 Venn diagrams of genes differentially expressed > 2-fold in iron-limited conditions compared to iron-rich conditions in COM1C2 and DTXR. A. Genes up- regulated (red) and down-regulated (green) in COM1 and DTXR were compared. Numbers in the circles indicated the number of genes regulated by iron; the outer circles represent genes in

COM1 and the inner circles represent genes in DTXR.

118

119

CHAPTER 5

CHARACTERIZATION OF THE PUTATIVE IRON PERMEASE FTR1 DELETION

MUTANT

5.1 Introduction to Ftr1 iron permease

A homolog of the high-affinity iron uptake permease FTR1 was found in the P. furiosus genome, encoded by PF0723 (PfFtr1). The best characterized member of the Ftr1 family is Ftr1p from Saccharomyces cerevisiae, which shares 27% identity and 49% similarity with PfFtr1. The

Ftr1-dependent iron acquisition system in S. cerevisiae typically involves reduction of ferric iron by the extracellular iron reductase Fre [194]. Ferrous iron is then acquired by the multicopper oxidase Fet3 which complexes with Ftr1 and reoxidizes Fe(II) to Fe(III) and transports iron into cells [101]. Ftr1 are also common in bacteria, being found mainly in proteobacteria, actinobacteria, cyanobacteria and firmicutes (Fig. 5.1). In bacteria, only the cyanobacteria appear to possess Ftr1p-encoding genes that are associated with genes specifying Fet3p-like proteins.

Other bacteria contain genes encoding different types of „iron-related‟ proteins adjacent to the ftr1 gene, but these are clearly distinct from Fet3p. As an example, the Ftr1 homolog-encoding gene in E. coli (efeU) is part of an iron- and Fur-repressed tri-cistronic operon, efeUOB [104].

Studies with EfeU suggest a role in ferrous iron transport. EfeB is a heme-containing periplasmic protein and has a low pH-enhanced peroxidase activity [195]. The specific roles of EfeO and

EfeB are unclear, but efeUOB-like operons can be found in many other bacteria, including B. subtilis [196] and N. meningitidis [197], suggesting that they have a potential role in iron uptake.

120

PfFtr1 shares 38% identity and 56% similarity with the EfeU of E. coli. A conserved five-amino acid motif REGLE for Ftr1-family proteins was found in PfFtr1 sequence and in homologs in Thermococcaceae species including T. barophilus, T. onnurineus NA1, T. sibiricus

MM 739 and P. yayanosii CH1 (Fig. 5.1). Transmembrane helix analysis predicts eight transmembrane helices, similar to the topology predicted for EfeU (Fig. 5.1). No homologs of

Fet3 and Fre are found in P. furiosus. Considering that the environment in which P. furiosus grows is reductive and most of the iron might be in the ferrous form, it is unlikely that the organism has a ferric iron uptake system as in yeast. The ORFs adjacent to ftr1 (PF0723) are alkyl hydroperoxide reductase I (PF0722) and a Trx-like protein (PF0724), both of which are cytoplasmic and likely involved in the defense of oxidative stress. Since ftr1 is significantly up- regulated according to microarray analysis (Table 4.1) while these adjacent genes are not, the chance of the genes being in an operon is low. The significant up-regulation of ftr1 during iron limitation also suggests a potential role in iron uptake in P. furiosus.

5.2 Characterization of the ftr1 deletion mutant

5.2.1 Construction of ftr1 deletion strain

The Δftr1 mutant, designated as FTR1, was supplied by Dr. Gina Lipscomb of the

University of Georgia. In brief, it was constructed using a vector containing 1 kb flanking regions of ftr1 on either side of the PgdhpyrF cassette, which was then transformed into the naturally competent strain COM1 to make a markerless deletion. The vector was mixed with freshly prepared COM1 culture and plate on solid medium without uracil to let the transformation begin. The first round of selection was for crossover event 1, in which the vector containing the PgdhpyrF cassette was integrated into COM1 genome and resulting in a strain that was uracil prototrophic. Colonies were isolated, grown and plated for the selection of crossover

121 event 2, in which the vector was removed from the genome by homologous recombination and left a fifty-percent chance for the generation of ftr1 deletion strain (Fig. 5.2). PCR and sequence analyses were used to confirm gene deletion in FTR1. Using Dr. Lipscomb‟s strain, RNA was isolated for qPCR and the result was compared to that of wild type. It indicated that ftr1 was detected for wild type around 16th cycle of PCR, while in FTR1 mutant the amplification curve only started to increase after 35th cycle, possibly due to non-specific amplification of PCR products. Hence, it was concluded that FTR1 is a clean deletion strain.

5.2.2 Effect of iron availability on FTR1 growth

Growth of FTR1 along with COM1, which is the parent strain used to construct FTR1, and wild type strains were compared after three consecutive transfers with inocula at mid- exponential phase (~5 x107 cells/mL) in an iron-limited maltose-based medium. For FTR1 and

COM1, 40 μM uracil was supplemented to the medium to assure growth.

FTR1, COM1 and wild type were grown in iron-limited medium. Samples were taken hourly from growing cultures of the three strains to give growth curves where cell density was represented by protein concentrations. Contrary to expectations, no obvious growth phenotype was observed for FTR1, suggesting that even though ftr1 is the most up-regulated gene in iron- limited conditions, single deletion of this iron permease is not sufficient to cause cellular iron deficiency when the medium is iron-limited (Fig. 5.3). Growth of FTR1 and COM1 was also measured in YEM-based medium without any trace minerals in the stock solution, including iron, zinc, copper, manganese, molybdenum, aluminum, cobalt and nickel. Once more growth of

COM1 and FTR1 appeared impaired compared to growth in original YEM, but no significant difference was found between the two strains.

122

5.2.3 Intracellular iron content in FTR1

To determine the overall intracellular concentration of iron present in the FTR1 and

COM1 strains, ICP-MS measurements of total cell fractions were carried out with cells cultured under different growth conditions and at different growth phase. Samples were harvested at the same time point during growth for both strains and growth conditions. Differences in intracellular iron levels were detected in strains under iron-rich and iron-limited conditions in exponential and stationary phase (Fig. 5.4). Cultures grown in iron-rich medium had a higher concentration of intracellular iron in exponential phase compared to stationary phase, indicating a steady accumulation of cellular iron when medium iron is sufficient for growth. In iron-limited medium, however, intracellular iron concentration remained unchanged throughout the growth phase and was similar to that in iron-rich medium at exponential phase. Apparently, the lack of iron in the medium was insufficient to support cell growth, so cells were forced to stop iron uptake and eventually enter stationary phase. However, the intracellular iron concentrations in

FTR1 were similar to those in COM1 under all conditions.

It is possible that iron uptake in P. furiosus is satisfied mainly by constitutively expressed iron transporters and that the up-regulation of Ftr1 does not play a major role even during iron limitation. Alternatively, deletion of Ftr1 may be compensated by the up-regulation of other transporters. Further studies are required to determine the role of Ftr1 in iron uptake by P. furiosus.

123

Figure 5.1 Analysis of PfFtr1 protein sequence. A. Phylogenetic distribution of Ftr1 family proteins. Ftr1 is conserved in phylogenetically diverse bacteria, including proteobacteria, actinobacteria, cyanobacteria and firmicutes. The phylogenetic tree was generated at the

STRING site (http://string.embl.de/) using orthology information from the COG database. B.

Sequence alignment of PfFtr1 and EfeU from E. coli using BLAST. The conserved five-amino acid motif REGLE essential for iron uptake [101] is in red. Transmembrane helix prediction (C) for PfFtr1 and (D) EfeU using TMHMM 2.0 (http://www.cbs.dtu.dk/services/TMHMM/).

Predict transmembrane helices in red, intracellular loop in blue and extracellular loop in pink. E.

Ftr1 homolog conserved in Thermococcaceae species.

124

125

Score = 142 bits (359), Expect = 4e-42, Method: Compositional matrix adjust.

Identities = 101/268 (38%), Positives = 150/268 (56%), Gaps = 17/268 (6%)

PfFtr1 MIGQFLITFREALEAAIIVAIIIAYLKRTDREEQIRDVWIGVGLSVLASVVLGAIILGLY 60 M FLI RE LEAA+IV++I +YLKRT R I +WIGV L+ + + LG I efeU MFVPFLIMLREGLEAALIVSLIASYLKRTQRGRWIGVMWIGVLLAAVLCLGLGIFINETT 60

PfFtr1 GGI--EEKELFEGIASYLAVIVLTSMIYWMATKGRNIRAEIERKVSKAI-----NPLALV 113 G +E+ELFEGI + +AV++LT M++WM RN++ ++E+ V A+ + ALV efeU GEFPQKEQELFEGIVAVIAVVILTWMVFWMRKVSRNVKVQLEQAVDSALQRGNHHGWALV 120

PfFtr1 GFTFIVVFREGLETVLFLTPFAIQDLG-GTLMGLVSGIIGALILAYLIYGVGMRINLRTF 172 F V REGLE+V FL QD+G +G + G+ A++L +L+Y G+R+NL F efeU MMVFFAVAREGLESVFFLLAAFQQDVGIWPPLGAMLGLATAVVLGFLLYWGGIRLNLGAF 180

PfFtr1 FYYSSILLVFVAAGLAGYGTHELIEWAEEEGMNLGFIEETAYNLGIPEDSVFHHKGVIGS 232 F ++S+ ++FVAAGLA E G+ F +E A+++ +V + G+ efeU FKWTSLFILFVAAGLAAGAIRAF----HEAGLWNHF-QEIAFDM----SAVLSTHSLFGT 231

PfFtr1 IFAVLFGYSVKMEWGRVIVQFGYLIVTL 260 + +FGY V V F YLI L efeU LMEGIFGYQEAPSVSEVAVWFIYLIPAL 259

126

127

Figure 5.2 Construction and verification of the ftr1 deletion strain. A. The knockout vector was constructed with 1 kb ftr1 gene flanking regions and the PgdhpyrF cassette for prototrophic selection. The knockout vector was transformed to the natural competent strain COM1 (ΔpyrF); the first crossover event resulted in the integration of the vector into COM1 genome and made the strain uracil prototrophic; in the second crossover event the vector was removed from genome DNA, and the resulting strains would be either the original COM1 or the strain with ftr1 deletion. B. Colonies from the second crossover event were picked and genomic DNA was isolated. Primers were designed to amplify the region containing the 1 kb ftr1 flanking region.

Agarose DNA gel was performed to confirm their length, as the PCR product from the fur deletion strain should be shorter than the original sequence, indicated as red asterisk. All the experiments above were performed by Dr. Gina Lipscomb.

128

129

Figure 5.3 Growth of wild type, COM1 and FTR1 in an iron-limited medium. Growth curves of wild type (blue), COM1 (red) and FTR1 (purple) in an iron-limited medium. Inocula for each culture were passed three times in YEM-Fe to remove the residual iron associated with the cells. All glassware was washed with 2% nitric acid overnight. Protein concentrations of cell extracts were measured using the Bradford assay.

130

25.0 /ml)

20.0

ug (

15.0

10.0 WT Cell Culture Culture Cell COM1

5.0 FTR1 Protein / Protein 0.0 0 2 4 6 8 10 12 14 Time (hour)

131

Figure 5.4 Intracellular iron content of COM1 and FTR1. Cultures of COM1 and FTR1 grown in iron-limited medium were harvested at exponential-phase or stationary phase. Cell pellets were washed three times with 1x base salt solution to remove extracellular traces of salts and lysed using Lysis Buffer containing 20 mM sodium phosphate, pH 7.4, made with ultrapure water was added at ~1 g cells per 3 mL lysis buffer. Lysate were then ultracentrifuged and supernatant was collected for the measurement of intracellular iron contents using ICP-MS (See

Section 2.9 for details). The iron contents measured by ICP-MS were normalized to the protein concentration of each sample for the analysis.

132

Fe (nmol/ug protein) COM1 FTR1 5.01E-02 4.55E-02

2.62E-02 2.56E-02 1.99E-02 1.90E-02 1.61E-02 1.61E-02

CMM LOG CMM-Fe LOG CMM STAT CMM-Fe STAT

133

CHAPTER 6

CONCLUSIONS

6.1 Effect of iron on the growth of Pyrococcus furiosus

Since P. furiosus is an aquatic anaerobic, S0-reducing hyperthermophilic archaeon that was isolated from a marine hydrothermal vent, the solubility of iron in its environment and the degree to which iron is accessible is largely unknown, especially since sulfide, the product of S0 reduction, complexes with Fe(II) to form insoluble FeS [112].

A widespread transcriptional response induced by iron was observed in P. furiosus; genes regulated include those encoding iron transporters (PF0723 and PF0858-PF0857), an iron- containing protein (PF0346), oxidative stress protection proteins (PF1033 and PF0692), proteins involved in amino acid metabolism (PF0935-0942 and PF1657-1666) and a number of hypothetical proteins. Unlike most aerobic bacteria, which contain multiple copies of genes encoding iron transport systems [114, 198], genes involved in iron acquisition in P. furiosus are more limited (Table 1.1). Genes involved in siderophore production were not found in P. furiosus, which might be due to the fact that the environment in which the organism survives is slightly acidic and reductive so that the iron pool mostly remains in its reduced form. The presence of putative ABC-type transporters in its genome indicates that P. furiosus might simply steal the iron-complexed siderophores produced by other bacteria [199].

Under iron-limited conditions, growth of P. furiosus reaches stationary phase earlier and at a lower cell density than growth in iron-rich conditions. This may be due to the lack of iron in the medium causing some of the cells to lyse and to release the intracellular iron to compensate

134 the growth of others. Otherwise the growth rate in iron-rich and iron-limited media appears to be the same (Fig. 4.16). It was confirmed by measuring the intracellular iron content that similar amounts of iron accumulate in cells grown under iron-rich and iron-limited conditions during the exponential growth phase, but the iron content of cells at stationary phase is significantly higher under iron-rich conditions indicating continuing iron uptake and storage when environmental iron is in abundance (Fig. 5.4).

Deletion of the most up-regulated iron transporter under iron-limited conditions ftr1, failed to induce any growth phenotype in iron-limited medium, suggesting that the primary iron transport system might be constitutively expressed in P. furiosus. It was shown by the microarray data that PF1774, the gene encoding a component of an ABC-type iron transporter, along with genes encoding the S-layer protein and SOR, is one of the most highly expressed genes in P. furiosus, indicating a potential role in iron acquisition. A similar stress response was observed for P. furiosus in oxidative shock, whereby the major oxidative stress machinery consisting of

PF1281 (SOR), PF1282 (rubredoxin) and PF1283 (rubrerythrin), is constitutively expressed at a high level rather than induced by oxidative stress [94].

6.2 PfDtxR-DNA binding affinity

PfDtxR is likely to have a small regulon. The comparison of the transcriptional regulation in COM1C2 and DTXR in response to iron revealed a potential PfDtxR regulon consisting of a limited number of genes (Table 4.1 and 4.2, shaded in grey). DNA-binding affinity has only been characterized for the promoters of ftr1 and feoAB so far. Therefore, the regulatory role of PfDtxR on the rest of the genes remains unknown. However, no conserved motifs were found within the promoters of genes potentially regulated by PfDtxR using MEME [185], indicating the genes might not all be directly regulated by PfDtxR. A motif search within a database of ORF upstream

135 sequences (UORs) extracted from the P. furiosus genome (software developed by Dr. Darin

Cowart, University of Georgia) revealed that the palindrome DNA sequence CCTAAn5TTAGG predicted to be the PfDtxR binding site, was only found in the ftr1 promoter. Furthermore, even though the specific binding of PfDtxR to the promoters of the two iron transporters has been confirmed, the EMSA of PfDtxR with an artificial DNA library showed no binding shift (data not shown). It is possibly due to the protein‟s high DNA-binding specificity that only a small pool of DNA from the artificial library can be recognized, hence the binding cannot be observed in EMSA.

DtxR functions as a repressor for the iron transporters in the presence of iron in P. furiosus (see below). PfDtxR binds to DNA in the EMSA without any metal ion added, potentially suggesting that Fe2+ or other metal ions (e.g., Zn2+) are already bound to the protein as it was purified [200, 201]. Metal analysis needs to be performed to determine what metal

PfDtxR is associated with.

6.3 Physiology of DTXR mutant and the mode of regulation by DtxR

PfDtxR binds specifically to the promoters of genes encoding the putative iron transporters ftr1 (PF0723) and feoAB (PF0857-58). Transcriptions of these genes are repressed in the control strain COM1C2 in iron-rich media, but no longer regulated in DTXR strain in response to iron. These results indicate the role of DtxR as an iron-dependent repressor for the iron transporters in P. furiosus. In contrast to most aerobic bacteria, in which a global transcriptional response to iron is observed [114, 196], the microarray analysis shows a relatively small number of genes that are potentially regulated by PfDtxR in response to iron. A similar effect has been reported for T. kodakarensis, in which the expression of two feoAB operons and a putative ZIP family transporter are down-regulated in wild type compared to the ΔdtxR strain

136 under iron-rich conditions [151]. Similar effects have also been described for the Δfur mutants of the obligate anaerobic bacteria Desulfovibrio vulgaris, Dichelobacter nodosus and Coxiella burnetii in which, under anaerobic growth, only a limited number of genes are differentially expressed in response to iron and no growth phenotype is observed in the deletion mutants [187,

202, 203]. In these organisms, Fur acts as a repressor of genes encoding iron transporters, the expression of which are derepressed in the deletion mutant. Yet, the derepressed iron acquisition systems of the deletion mutants fail to offer any advantage to the growth in iron-limited media or to introduce any disadvantage to the growth in iron-rich media. On the other hand, defects in growth were observed for several fur mutants of aerobic bacteria [200, 204, 205].

Ferrous iron is the catalyst of Fenton reaction, which results in the production of extremely reactive oxygen species (ROS) that leads to damage of biological macromolecules and cell death. Therefore, iron metabolism in aerobic bacteria is usually under strict regulation to protect cells against oxidative stress; the absence of proper control over iron metabolism could cause difficulties in coping with oxidative stress, not only because the chances of iron-induced

Fenton reaction, but also because many of the oxidative stress enzymes require iron as a cofactor

[205-207]. Anaerobic organisms, in contrast, are much less threatened by oxidative stress and ramification of the Fenton reaction since they do not experience oxygen under typical growth conditions. Therefore, it is possible that P. furiosus constitutively expresses its major iron acquisition system; however, further research needs to be performed to confirm this hypothesis and to better understand the iron metabolism in P. furiosus.

137

REFERENCES

1. Woese, C.R. and G.E. Fox, Phylogenetic structure of the prokaryotic domain: the

primary kingdoms. Proc Natl Acad Sci U S A, 1977. 74(11): p. 5088-90.

2. Woese, C.R. and G.E. Fox, The concept of cellular evolution. Journal of Molecular

Evolution, 1977. 10(1): p. 1-6.

3. Sapp, J., Two faces of the prokaryote concept. Int Microbiol, 2006. 9(3): p. 163-72.

4. Graham, D.E., et al., An archaeal genomic signature. Proc Natl Acad Sci U S A, 2000.

97(7): p. 3304-8.

5. Gaasterland, T., Archaeal genomics. Current Opinion in Microbiology, 1999. 2(5): p.

542-7.

6. Fournier, G.P., et al., Evolution of the archaea: emerging views on origins and phylogeny.

Res Microbiol, 2011. 162(1): p. 92-98.

7. Chaban, B., S.Y. Ng, and K.F. Jarrell, Archaeal habitats--from the extreme to the

ordinary. Canadian Journal of Microbiology, 2006. 52(2): p. 73-116.

8. Robertson, C.E., et al., Phylogenetic diversity and ecology of environmental Archaea.

Current Opinion in Microbiology, 2005. 8(6): p. 638-42.

9. Allers, T. and M. Mevarech, Archaeal genetics - the third way. Nat Rev Genet, 2005.

6(1): p. 58-73.

10. Zillig, W., K.O. Stetter, and D. Janekovic, DNA-dependent RNA polymerase from the

archaebacterium Sulfolobus acidocaldarius. Eur J Biochem, 1979. 96(3): p. 597-604.

138

11. Nikolov, D.B., et al., Crystal structure of a TFIIB-TBP-TATA-element ternary complex.

Nature, 1995. 377(6545): p. 119-28.

12. Aravind, L. and E.V. Koonin, DNA-binding proteins and evolution of transcription

regulation in the archaea. Nucleic Acids Research, 1999. 27(23): p. 4658-70.

13. Bell, S.D. and S.P. Jackson, Transcription and translation in Archaea: a mosaic of

eukaryal and bacterial features. Trends Microbiol, 1998. 6(6): p. 222-8.

14. Zillig, W., K.O. Stetter, and M. Tobien, DNA-dependent RNA polymerase from

Halobacterium halobium. Eur J Biochem, 1978. 91(1): p. 193-9.

15. Langer D, H.J., Thuriaux P and Zillig W., Transcription in archaea: similarity to that in

eucarya. Proc Natl Acad Sci USA, 1995. 92: p. 5768–5772.

16. Brinkman, A.B., et al., The Lrp family of transcriptional regulators. Molecular

Microbiology, 2003. 48(2): p. 287-94.

17. Ouhammouch, M., Transcriptional regulation in Archaea. Curr Opin Genet Dev, 2004.

14(2): p. 133-8.

18. Armache, K.J., et al., Structures of complete RNA polymerase II and its subcomplex,

Rpb4/7. Journal of Biological Chemistry, 2005. 280(8): p. 7131-4.

19. Huet, J., et al., Archaebacteria and eukaryotes possess DNA-dependent RNA polymerases

of a common type. Embo Journal, 1983. 2(8): p. 1291-4.

20. Darst, S.A., Bacterial RNA polymerase. Curr Opin Struct Biol, 2001. 11(2): p. 155-62.

21. Hirata, A., B.J. Klein, and K.S. Murakami, The X-ray crystal structure of RNA

polymerase from Archaea. Nature, 2008. 451(7180): p. 851-4.

22. Hirata, A. and K.S. Murakami, Archaeal RNA polymerase. Curr Opin Struct Biol, 2009.

19(6): p. 724-31.

139

23. Korkhin, Y., et al., Evolution of Complex RNA Polymerases: The Complete Archaeal

RNA Polymerase Structure. Plos Biology, 2009. 7(5): p. e102.

24. Kusser, A.G., et al., Structure of an archaeal RNA polymerase. Journal of Molecular

Biology, 2008. 376(2): p. 303-7.

25. Kuhn, C.D., et al., Functional architecture of RNA polymerase I. Cell, 2007. 131(7): p.

1260-72.

26. Jasiak, A.J., et al., Structural biology of RNA polymerase III: subcomplex C17/25 X-ray

structure and 11 subunit enzyme model. Molecular Cell, 2006. 23(1): p. 71-81.

27. Beasley, F.C. and D.E. Heinrichs, Siderophore-mediated iron acquisition in the

staphylococci. J Inorg Biochem, 2010. 104(3): p. 282-288.

28. Butler, A. and R.M. Theisen, Iron(III)-siderophore coordination chemistry: Reactivity of

marine siderophores. Coordination Chemistry Reviews, 2010. 254(3-4): p. 288-296.

29. Grigg, J.C., et al., Structural biology of heme binding in the Staphylococcus aureus Isd

system. J Inorg Biochem, 2010. 104(3): p. 341-348.

30. de Carvalho, C.C.C.R. and P. Fernandes, Production of Metabolites as Bacterial

Responses to the Marine Environment. Marine Drugs, 2010. 8(3): p. 705-727.

31. Kosa, P.F., et al., The 2.1-A crystal structure of an archaeal preinitiation complex:

TATA-box-binding protein/transcription factor (II)B core/TATA-box. Proc Natl Acad Sci

U S A, 1997. 94(12): p. 6042-7.

32. Kim, J.L. and S.K. Burley, 1.9 A resolution refined structure of TBP recognizing the

minor groove of TATAAAAG. Nature Structural Biology, 1994. 1(9): p. 638-53.

33. Nikolov, D.B., et al., Crystal structure of a human TATA box-binding protein/TATA

element complex. Proc Natl Acad Sci U S A, 1996. 93(10): p. 4862-7.

140

34. Bell, S.D., et al., Orientation of the transcription preinitiation complex in archaea. Proc

Natl Acad Sci U S A, 1999. 96(24): p. 13662-7.

35. Littlefield, O., Y. Korkhin, and P.B. Sigler, The structural basis for the oriented assembly

of a TBP/TFB/promoter complex. Proc Natl Acad Sci U S A, 1999. 96(24): p. 13668-73.

36. Xie, Y. and J.N. Reeve, Transcription by Methanothermobacter thermautotrophicus RNA

polymerase in vitro releases archaeal transcription factor B but not TATA-box binding

protein from the template DNA. Journal of Bacteriology, 2004. 186(18): p. 6306-10.

37. Spitalny, P. and M. Thomm, Analysis of the open region and of DNA-protein contacts of

archaeal RNA polymerase transcription complexes during transition from initiation to

elongation. Journal of Biological Chemistry, 2003. 278(33): p. 30497-505.

38. Kostrewa, D., et al., RNA polymerase II-TFIIB structure and mechanism of transcription

initiation. Nature, 2009. 462(7271): p. 323-30.

39. Grunberg S, B.M., Naji S and Thomm M, Transcription factor E is a part of

transcription elongation complexes. J Biol Chem, 2007. 282: p. 35482-90.

40. M, T., Transcription: Mechanism and Regulation. In: Cavicchioli R. (ed.) Archaea:

Molecular and Cellular Biology. Washington, DC: ASM Press., 2007.

41. Bryson K, M.L., Marsden RL, Ward JJ, Sodhi JS. & Jones DT, Protein structure

prediction servers at University College London. Nucleic Acids Research, 2005. 33(Web

Server issue): p. W36-38.

42. Yamamoto, T., et al., Crystal structure of TBP-interacting protein (Tk-TIP26) and

implications for its inhibition mechanism of the interaction between TBP and TATA-DNA.

Protein Science, 2006. 15(1): p. 152-61.

141

43. Bell, S.D., Archaeal transcriptional regulation--variation on a bacterial theme? Trends

Microbiol, 2005. 13(6): p. 262-265.

44. Geiduschek, E.P.a.M.O., Archaeal transcription and its regulators. Mol Microbiol, 2005.

56(6): p. 1397-1407.

45. Bell, S.D., et al., Transcriptional regulation of an archaeal operon in vivo and in vitro.

Molecular Cell, 1999. 4(6): p. 971-82.

46. Brinkman, A.B., et al., An Lrp-like transcriptional regulator from the archaeon

Pyrococcus furiosus is negatively autoregulated. J Biol Chem, 2000. 275(49): p. 38160-9.

47. Enoru-Eta, J., et al., Purification and characterization of Sa-lrp, a DNA-binding protein

from the extreme thermoacidophilic archaeon Sulfolobus acidocaldarius homologous to

the bacterial global transcriptional regulator Lrp. J Bacteriol, 2000. 182(13): p. 3661-72.

48. Enoru-Eta, J., et al., High resolution contact probing of the Lrp-like DNA-binding protein

Ss-Lrp from the hyperthermoacidophilic crenarchaeote Sulfolobus solfataricus P2. Mol

Microbiol, 2002. 45(6): p. 1541-55.

49. Peeters, E., et al., Ss-LrpB, a novel Lrp-like regulator of Sulfolobus solfataricus P2, binds

cooperatively to three conserved targets in its own control region. Mol Microbiol, 2004.

54(2): p. 321-36.

50. Peeters, E., et al., Ss-LrpB from Sulfolobus solfataricus condenses about 100 base pairs

of its own operator DNA into globular nucleoprotein complexes. J Biol Chem, 2006.

281(17): p. 11721-8.

51. Bell, S.D.a.S.P.J., Mechanism of autoregulation by an archaeal transcriptional repressor.

J Biol Chem, 2000. 275(41): p. 31624-9.

142

52. Yokoyama, K., et al., Feast/famine regulatory proteins (FFRPs): Escherichia coli Lrp,

AsnC and related archaeal transcription factors. FEMS Microbiol Rev, 2006. 30(1): p.

89-108.

53. Brinkman, A.B., et al., The Sulfolobus solfataricus Lrp-like protein LysM regulates lysine

biosynthesis in response to lysine availability. J Biol Chem, 2002. 277(33): p. 29537-49.

54. Lee, S.J., et al., TrmB, a sugar-specific transcriptional regulator of the trehalose/maltose

ABC transporter from the hyperthermophilic archaeon Thermococcus litoralis. J Biol

Chem, 2003. 278(2): p. 983-90.

55. Fitsanakis, V.A., et al., Manganese (Mn) and Iron (Fe): Interdependency of Transport

and Regulation. Neurotoxicity Research, 2010. 18(2): p. 124-131.

56. Krug, M., et al., Crystal structure of the sugar binding domain of the archaeal

transcriptional regulator TrmB. J Biol Chem, 2006. 281(16): p. 10976-82.

57. Lee, S.J., et al., Characterization of the TrmB-like protein, PF0124, a TGM-recognizing

global transcriptional regulator of the hyperthermophilic archaeon Pyrococcus furiosus.

Mol Microbiol, 2007. 65(2): p. 305-18.

58. Kanai, T., et al., A global transcriptional regulator in Thermococcus kodakaraensis

controls the expression levels of both glycolytic and gluconeogenic enzyme-encoding

genes. J Biol Chem, 2007. 282(46): p. 33659-70.

59. Maruyama, H., et al., Histone and TK0471/TrmBL2 form a novel heterogeneous genome

architecture in the hyperthermophilic archaeon Thermococcus kodakarensis. Mol Biol

Cell, 2011. 22(3): p. 386-98.

60. Lie, T.J.a.J.A.L., A novel repressor of nif and glnA expression in the methanogenic

archaeon Methanococcus maripaludis. Mol Microbiol, 2003. 47(1): p. 235-46.

143

61. Lie, T.J., G.E. Wood, and J.A. Leigh, Regulation of nif expression in Methanococcus

maripaludis: roles of the euryarchaeal repressor NrpR, 2-oxoglutarate, and two perators.

J Biol Chem, 2005. 280(7): p. 5236-41.

62. Plosser, P.a.F.P., A bZIP protein from halophilic archaea: structural features and dimer

formation of cGvpE from Halobacterium salinarum. Mol Microbiol, 2002. 45(2): p. 511-

20.

63. Pfeifer, F., et al., A p-loop motif and two basic regions in the regulatory protein GvpD

are important for the repression of gas vesicle formation in the archaeon Haloferax

mediterranei. Microbiology, 2001. 147(Pt 1): p. 63-73.

64. Zimmermann, P.a.F.P., Regulation of the expression of gas vesicle genes in Haloferax

mediterranei: interaction of the two regulatory proteins GvpD and GvpE. Mol Microbiol,

2003. 49(3): p. 783-94.

65. Vierke, G., et al., A novel archaeal transcriptional regulator of heat shock response. J

Biol Chem, 2003. 278(1): p. 18-26.

66. Liu, W., et al., Crystal structure of the archaeal heat shock regulator from Pyrococcus

furiosus: a molecular chimera representing eukaryal and bacterial features. J Mol Biol,

2007. 369(2): p. 474-88.

67. M. Ouhammouch, G.E.L., W. Hausner, A. J. Simpson, N. M. A. El-Sayed, E. P.

Geiduschek, Promoter architecture and response to a positive regulator of archaeal

transcription. Mol Microbiol, 2005. 56(3): p. 625-637.

68. Ouhammouch, M., W. Hausner, and E.P. Geiduschek, TBP domain symmetry in basal

and activated archaeal transcription. Mol Microbiol, 2009. 71(1): p. 123-31.

144

69. Kessler, A., et al., A novel archaeal regulatory protein, Sta1, activates transcription from

viral promoters. Nucleic Acids Research, 2006. 32(17): p. 4837-45.

70. Hotta, K., et al., Siderophore-mediated iron acquisition in Bacillus anthracis and related

strains. Microbiology-Sgm, 2010. 156: p. 1918-1925.

71. d'Hardemare, A.D., G. Serratrice, and J.L. Pierre, Synthetic Approaches for Iron Type

Chelators. Mini-Reviews in Organic Chemistry, 2010. 7(2): p. 125-133.

72. Stetter, K.O., Hyperthermophiles in the history of life. Philos Trans R Soc Lond B Biol

Sci, 2006. 361(1474): p. 1837-42; discussion 1842-3.

73. Fiala, G., and Stetter, K.O., Pyrococcus-Furiosus Sp-Nov Represents a Novel Genus of

Marine Heterotrophic Archaebacteria Growing Optimally at 100-Degrees C. Archives of

Microbiology 1986. 145: p. 56-61.

74. Cvetkovic, A., et al., Microbial metalloproteomes are largely uncharacterized. Nature,

2010. 466(7307): p. 779-U18.

75. Mukund, S. and M.W.W. Adams, The Novel Tungsten-Iron-Sulfur Protein of the

Hyperthermophilic Archaebacterium, Pyrococcus-Furiosus, Is an Aldehyde Ferredoxin

Oxidoreductase - Evidence for Its Participation in a Unique Glycolytic Pathway. Journal

of Biological Chemistry, 1991. 266(22): p. 14208-14216.

76. Bryant, F.O. and M.W.W. Adams, Characterization of Hydrogenase from the

Hyperthermophilic Archaebacterium, Pyrococcus-Furiosus. Journal of Biological

Chemistry, 1989. 264(9): p. 5070-5079.

77. Tatur, J., et al., A highly thermostable ferritin from the hyperthermophilic archaeal

anaerobe Pyrococcus furiosus. Extremophiles, 2006. 10(2): p. 139-148.

145

78. Mai, X.H. and M.W.W. Adams, Indolepyruvate Ferredoxin Oxidoreductase from the

Hyperthermophilic Archaeon Pyrococcus-Furiosus - a New Enzyme Involved in Peptide

Fermentation. Journal of Biological Chemistry, 1994. 269(24): p. 16726-16732.

79. Mukund, S. and M.W.W. Adams, Glyceraldehyde-3-Phosphate Ferredoxin

Oxidoreductase, a Novel Tungsten-Containing Enzyme with a Potential Glycolytic Role

in the Hyperthermophilic Archaeon Pyrococcus-Furiosus. Journal of Biological

Chemistry, 1995. 270(15): p. 8389-8392.

80. Jenney, F.E., et al., Anaerobic microbes: Oxygen detoxification without superoxide

dismutase. Science, 1999. 286(5438): p. 306-309.

81. Roy, R., et al., Purification and molecular characterization of the tungsten-containing

formaldehyde ferredoxin oxidoreductase from the hyperthermophilic archaean

Pyrococcus furiosus: The third of a putative five-member tungstoenzyme family. Journal

of Bacteriology, 1999. 181(4): p. 1171-1180.

82. Heider, J., X.H. Mai, and M.W.W. Adams, Characterization of 2-ketoisovalerate

ferredoxin oxidoreductase, a new and reversible coenzyme A-dependent enzyme involved

in peptide fermentation by hyperthermophilic archaea. Journal of Bacteriology, 1996.

178(3): p. 780-787.

83. Blamey, J.M. and M.W.W. Adams, Purification and Characterization of Pyruvate

Ferredoxin Oxidoreductase from the Hyperthermophilic Archaeon Pyrococcus-Furiosus.

Biochimica Et Biophysica Acta, 1993. 1161(1): p. 19-27.

84. Ma, K.S., R. Weiss, and M.W.W. Adams, Characterization of hydrogenase II from the

hyperthermophilic archaeon Pyrococcus furiosus and assessment of its role in sulfur

reduction. Journal of Bacteriology, 2000. 182(7): p. 1864-1871.

146

85. Ma, K. and M.W.W. Adams, Sulfide Dehydrogenase from the Hyperthermophilic

Archaeon Pyrococcus-Furiosus - a New Multifunctional Enzyme Involved in the

Reduction of Elemental Sulfur. Journal of Bacteriology, 1994. 176(21): p. 6509-6517.

86. Weinberg, M.V., et al., Rubrerythrin from the hyperthermophilic archaeon Pyrococcus

furiosus is a rubredoxin-dependent, iron-containing peroxidase. Journal of Bacteriology,

2004. 186(23): p. 7888-7895.

87. Roy, R. and M.W.W. Adams, Characterization of a fourth tungsten-containing enzyme

from the hyperthermophilic archaeon Pyrococcus furiosus. Journal of Bacteriology, 2002.

184(24): p. 6952-6956.

88. Hagen, W.R., et al., Novel structure and redox chemistry of the prosthetic groups of the

iron-sulfur flavoprotein sulfide dehydrogenase from Pyrococcus furiosus; evidence for a

[2Fe-2S] cluster with Asp(Cys)(3) ligands. Journal of Biological Inorganic Chemistry,

2000. 5(4): p. 527-534.

89. Aono, S., F.O. Bryant, and M.W.W. Adams, A Novel and Remarkably Thermostable

Ferredoxin from the Hyperthermophilic Archaebacterium Pyrococcus-Furiosus. Journal

of Bacteriology, 1989. 171(6): p. 3433-3439.

90. Bevers, L.E., et al., WOR5, a novel tungsten-containing aldehyde oxidoreductase from

Pyrococcus furiosus with a broad substrate specificity. Journal of Bacteriology, 2005.

187(20): p. 7056-7061.

91. Sapra, R., M.F.J.M. Verhagen, and M.W.W. Adams, Purification and characterization of

a membrane-bound hydrogenase from the hyperthermophilic archaeon Pyrococcus

furiosus. Journal of Bacteriology, 2000. 182(12): p. 3423-3428.

147

92. Tatur, J., W.R. Hagen, and P.M. Matias, Crystal structure of the ferritin from the

hyperthermophilic archaeal anaerobe Pyrococcus furiosus. Journal of Biological

Inorganic Chemistry, 2007. 12(5): p. 615-30.

93. Ramsay, B., et al., Dps-like protein from the hyperthermophilic archaeon Pyrococcus

furiosus. J Inorg Biochem, 2006. 100(5-6): p. 1061-8.

94. Strand, K.R., et al., Oxidative stress protection and the repair response to hydrogen

peroxide in the hyperthermophilic archaeon Pyrococcus furiosus and in related species.

Archives of Microbiology, 2010. 192(6): p. 447-59.

95. Williams, E., et al., Microarray analysis of the hyperthermophilic archaeon Pyrococcus

furiosus exposed to gamma irradiation. Extremophiles, 2007. 11(1): p. 19-29.

96. Schut, G.J., S.L. Bridger, and M.W. Adams, Insights into the metabolism of elemental

sulfur by the hyperthermophilic archaeon Pyrococcus furiosus: characterization of a

coenzyme A- dependent NAD(P)H sulfur oxidoreductase. Journal of Bacteriology, 2007.

189(12): p. 4431-41.

97. Clarkson, S.M., et al., The elemental sulfur-responsive protein (SipA) from the

hyperthermophilic archaeon Pyrococcus furiosus is regulated by sulfide in an iron-

dependent manner. Journal of Bacteriology, 2010. 192(21): p. 5841-3.

98. Dashper, S.G., et al., A novel Porphyromonas gingivalis FeoB plays a role in manganese

accumulation. Journal of Biological Chemistry, 2005. 280(30): p. 28095-102.

99. Kammler, M., C. Schon, and K. Hantke, Characterization of the ferrous iron uptake

system of Escherichia coli. Journal of Bacteriology, 1993. 175(19): p. 6212-9.

148

100. Velayudhan, J., et al., Iron acquisition and virulence in Helicobacter pylori: a major role

for FeoB, a high-affinity ferrous iron transporter. Molecular Microbiology, 2000. 37(2):

p. 274-86.

101. Stearman, R., et al., A permease-oxidase complex involved in high-affinity iron uptake in

yeast. Science, 1996. 271(5255): p. 1552-7.

102. Grosse, C., et al., A new ferrous iron-uptake transporter, EfeU (YcdN), from Escherichia

coli. Molecular Microbiology, 2006. 62(1): p. 120-31.

103. Rajasekaran, M.B., et al., EfeO-cupredoxins: major new members of the cupredoxin

superfamily with roles in bacterial iron transport. Biometals, 2010. 23(1): p. 1-17.

104. Cao, J., et al., EfeUOB (YcdNOB) is a tripartite, acid-induced and CpxAR-regulated,

low-pH Fe2+ transporter that is cryptic in Escherichia coli K-12 but functional in E. coli

O157:H7. Molecular Microbiology, 2007. 65(4): p. 857-75.

105. Johnston, A.W., et al., Living without Fur: the subtlety and complexity of iron-responsive

gene regulation in the symbiotic bacterium Rhizobium and other alpha-proteobacteria.

Biometals, 2007. 20(3-4): p. 501-11.

106. Helmann, J.D. and J.W. Lee, Functional specialization within the Fur family of

metalloregulators. Biometals, 2007. 20(3-4): p. 485-499.

107. Guedon, E. and J.D. Helmann, Origins of metal ion selectivity in the DtxR/MntR family of

metalloregulators. Molecular Microbiology, 2003. 48(2): p. 495-506.

108. Nobles, C.L. and A.W. Maresso, The theft of host heme by Gram-positive pathogenic

bacteria. Metallomics, 2011. 3(8): p. 788-796.

109. Ferry, J.G., The chemical biology of methanogenesis. Planetary and Space Science, 2010.

58(14-15): p. 1775-1783.

149

110. Chu, B.C., et al., Siderophore uptake in bacteria and the battle for iron with the host; a

bird's eye view. Biometals, 2010. 23(4): p. 601-611.

111. Fried, M., and Crothers, D.M., Equilibria and kinetics of lac repressor-operator

interactions by polyacrylamide gel electrophoresis. Nucleic Acids Research, 1981. 9: p.

6505-6525.

112. Adams, M.W., J. F. Holden, A. L. Menon, G. J. Schut, A. M. Grunden, C. Hou, A. M.

Hutchins, F. E. Jenney, Jr., C. Kim, K. Ma, G. Pan, R. Roy, R. Sapra, S. V. Story, and M.

F. Verhagen., Key role for sulfur in peptide metabolism and in regulation of three

hydrogenases in the hyperthermophilic archaeon Pyrococcus furiosus. J Bacteriol 2001.

183: p. 716-24.

113. Lipscomb, G.L., et al., Natural Competence in the Hyperthermophilic Archaeon

Pyrococcus furiosus Facilitates Genetic Manipulation: Construction of Markerless

Deletions of Genes Encoding the Two Cytoplasmic Hydrogenases. Applied and

Environmental Microbiology, 2011. 77(7): p. 2232-2238.

114. Andrews, S.C., et al., Global iron-dependent gene regulation in Escherichia coli - A new

mechanism for iron homeostasis. Journal of Biological Chemistry, 2003. 278(32): p.

29478-29486.

115. Kusters, J.G., et al., Helicobacter pylori ribBA-mediated riboflavin production is involved

in tron acquisition. Journal of Bacteriology, 1998. 180(6): p. 1473-1479.

116. Calderwood, S.B., P.I. Watnick, and J.R. Butterton, The interaction of the Vibrio

cholerae transcription factors, Fur and IrgB, with the overlapping promoters of two

virulence genes, irgA and irgB. Gene, 1998. 209(1-2): p. 65-70.

150

117. Torres, A.G. and S.M. Payne, Haem iron-transport system in enterohaemorrhagic

Escherichia coli O157:H7. Molecular Microbiology, 1997. 23(4): p. 825-833.

118. van Vliet, A.H., et al., Differential regulation of amidase- and formamidase-mediated

ammonia production by the Helicobacter pylori fur repressor. Journal of Biological

Chemistry, 2003. 278(11): p. 9052-7.

119. Masse, E. and S. Gottesman, A small RNA regulates the expression of genes involved in

iron metabolism in Escherichia coli. Proc Natl Acad Sci U S A, 2002. 99(7): p. 4620-5.

120. Masse, E., et al., Small RNAs controlling iron metabolism. Current Opinion in

Microbiology, 2007. 10(2): p. 140-5.

121. Mellin, J.R., et al., A novel fur- and iron-regulated small RNA, NrrF, is required for

indirect fur-mediated regulation of the sdhA and sdhC genes in Neisseria meningitidis.

Journal of Bacteriology, 2007. 189(10): p. 3686-94.

122. Wilderman, P.J., et al., Identification of tandem duplicate regulatory small RNAs in

Pseudomonas aeruginosa involved in iron homeostasis. Proc Natl Acad Sci U S A, 2004.

101(26): p. 9792-7.

123. Delany, I., R. Rappuoli, and V. Scarlato, Fur functions as an activator and as a repressor

of putative virulence genes in Neisseria meningitidis. Molecular Microbiology, 2004.

52(4): p. 1081-90.

124. Alamuri, P., et al., Regulation of the Helicobacter pylori Fe-S cluster synthesis protein

NifS by iron, oxidative stress conditions, and fur. Journal of Bacteriology, 2006. 188(14):

p. 5325-30.

151

125. Gancz, H., S. Censini, and D.S. Merrell, Iron and pH homeostasis intersect at the level of

Fur regulation in the gastric pathogen Helicobacter pylori. Infection and Immunity,

2006. 74(1): p. 602-14.

126. Ernst, F.D., et al., Transcriptional profiling of Helicobacter pylori Fur- and iron-

regulated gene expression. Microbiology, 2005. 151(Pt 2): p. 533-46.

127. Delany, I., et al., The Fur repressor controls transcription of iron-activated and -

repressed genes in Helicobacter pylori. Molecular Microbiology, 2001. 42(5): p. 1297-

309.

128. Delany, I., et al., Iron-dependent transcription of the frpB gene of Helicobacter pylori is

controlled by the Fur repressor protein. Journal of Bacteriology, 2001. 183(16): p. 4932-

129. Ernst, F.D., et al., Iron-responsive regulation of the Helicobacter pylori iron-cofactored

superoxide dismutase SodB is mediated by Fur. Journal of Bacteriology, 2005. 187(11): p.

3687-92.

130. Spiegelhalder, C., et al., Purification of Helicobacter pylori superoxide dismutase and

cloning and sequencing of the gene. Infection and Immunity, 1993. 61(12): p. 5315-25.

131. Lee, H.J., et al., Positive regulation of fur gene expression via direct interaction of fur in

a pathogenic bacterium, Vibrio vulnificus. Journal of Bacteriology, 2007. 189(7): p.

2629-36.

132. Small, S.K., et al., Positive control of ferric siderophore receptor gene expression by the

Irr protein in Bradyrhizobium japonicum. Journal of Bacteriology, 2009. 191(5): p. 1361-

152

133. Hantke, K. and S.I. Patzer, The ZnuABC high-affinity zinc uptake system and its regulator

zur in Escherichia coli. Molecular Microbiology, 1998. 28(6): p. 1199-1210.

134. O'Halloran, T.V., et al., Characterization of the metal receptor sites in Escherichia coli

Zur, an ultrasensitive zinc(II) metalloregulatory protein. Biochemistry, 2001. 40(35): p.

10417-10423.

135. Fabiano, E., et al., Sinorhizobium meliloti fur-like (Mur) protein binds a fur box-like

sequence present in the mntA promoter in a manganese-responsive manner. Applied and

Environmental Microbiology, 2007. 73(15): p. 4832-4838.

136. Johnston, A.W.B., et al., The Fur-like protein Mur of Rhizobium leguminosarum is a

Mn2+-responsive transcriptional regulator. Microbiology-Sgm, 2004. 150: p. 1447-1456.

137. Roe, J.H., et al., Nur, a nickel-responsive regulator of the Fur family, regulates

superoxide dismutases and nickel transport in Streptomyces coelicolor. Molecular

Microbiology, 2006. 59(6): p. 1848-1858.

138. Helmann, J.D. and J.W. Lee, The PerR transcription factor senses H2O2 by metal-

catalysed histidine oxidation. Nature, 2006. 440(7082): p. 363-367.

139. Helmann, J.D., et al., Regulation of the Bacillus subtilis fur and perR genes by PerR: Not

all members of the PerR regulon are peroxide inducible. Journal of Bacteriology, 2002.

184(12): p. 3276-3286.

140. Helmann, J.D., et al., Bacillus subtilis contains multiple Fur homologues: identification

of the iron uptake (Fur) and peroxide regulon (PerR) repressors. Molecular

Microbiology, 1998. 29(1): p. 189-198.

153

141. Pohl, E., et al., Architecture of a protein central to iron homeostasis: crystal structure

and spectroscopic analysis of the ferric uptake regulator. Molecular Microbiology, 2003.

47(4): p. 903-915.

142. Taylor, G.L. and M.A. Sheikh, Crystal structure of the Vibrio cholerae ferric uptake

regulator (Fur) reveals insights into metal co-ordination. Molecular Microbiology, 2009.

72(5): p. 1208-1220.

143. Dian, C., et al., The structure of the Helicobacter pylori ferric uptake regulator Fur

reveals three functional metal binding sites. Molecular Microbiology, 2011. 79(5): p.

1260-75.

144. Michaud-Soret, I., et al., Structural changes of Escherichia coli ferric uptake regulator

during metal-dependent dimerization and activation explored by NMR and x-ray

crystallography. Journal of Biological Chemistry, 2006. 281(30): p. 21286-21295.

145. Meyer-Klaucke, W., et al., Crystal structure and function of the zinc uptake regulator

FurB from Mycobacterium tuberculosis. Journal of Biological Chemistry, 2007. 282(13):

p. 9914-9922.

146. Duarte, V., et al., Crystal structure of the apo-PerR-Zn protein from Bacillus subtilis.

Molecular Microbiology, 2006. 61(5): p. 1211-1219.

147. Duarte, V., et al., Structural characterization of the active form of PerR: insights into the

metal-induced activation of PerR and Fur proteins for DNA binding. Molecular

Microbiology, 2009. 73(1): p. 20-31.

148. Duarte, V., et al., Structural and functional characterization of 2-oxo-histidine in

oxidized PerR protein. Nature Chemical Biology, 2009. 5(1): p. 53-59.

154

149. Roe, J.H., et al., Structural basis for the specialization of Nur, a nickel-specific Fur

homolog, in metal sensing and DNA recognition. Nucleic Acids Research, 2009. 37(10):

p. 3442-3451.

150. Marchler-Bauer, A., et al., CDD: a Conserved Domain Database for the functional

annotation of proteins. Nucleic Acids Research, 2011. 39: p. D225-D229.

151. Louvel, H., et al., The Fur iron regulator-like protein is cryptic in the hyperthermophilic

archaeon Thermococcus kodakaraensis. Fems Microbiology Letters, 2009. 295(1): p.

117-128.

152. De Lorenzo, V., L. Escolar, and J. Perez-Martin, Opening the iron box: Transcriptional

metalloregulation by the fur protein. Journal of Bacteriology, 1999. 181(20): p. 6223-

6229.

153. Hantke, K., Iron and metal regulation in bacteria. Current Opinion in Microbiology,

2001. 4(2): p. 172-177.

154. Lipscomb, G.L., et al., SurR: a transcriptional activator and repressor controlling

hydrogen and elemental sulphur metabolism in Pyrococcus furiosus. Molecular

Microbiology, 2009. 71(2): p. 332-49.

155. Grass, G., et al., A new ferrous iron-uptake transporter, EfeU (YcdN), from Escherichia

coli. Molecular Microbiology, 2006. 62(1): p. 120-131.

156. Jayaswal, R.K., et al., Transcriptomic Response of Listeria monocytogenes to Iron

Limitation and fur Mutation. Applied and Environmental Microbiology, 2010. 76(2): p.

406-416.

157. Pappenheimer, A.M., Jr, Studies in diphtheria toxin production. I. The effect of iron and

copper. Br J Exp Pathol, 1936. 17: p. 335-341.

155

158. Tao, X., et al., Iron, Dtxr, and the Regulation of Diphtheria-Toxin Expression. Molecular

Microbiology, 1994. 14(2): p. 191-197.

159. Schmitt, M.P., et al., Characterization of an iron-dependent regulatory protein (IdeR) of

Mycobacterium tuberculosis as a functional homolog of the diphtheria toxin repressor

(DtxR) from Corynebacterium diphtheriae. Infection and Immunity, 1995. 63(11): p.

4284-9.

160. Brune, I., et al., The DtxR protein acting as dual transcriptional regulator directs a

global regulatory network involved in iron metabolism of Corynebacterium glutamicum.

Bmc Genomics, 2006. 7: p. 21.

161. Wennerhold, J. and M. Bott, The DtxR regulon of Corynebacterium glutamicum. Journal

of Bacteriology, 2006. 188(8): p. 2907-18.

162. Horsburgh, M.J., et al., MntR modulates expression of the PerR regulon and superoxide

resistance in Staphylococcus aureus through control of manganese uptake. Molecular

Microbiology, 2002. 44(5): p. 1269-86.

163. Rodriguez, G.M., et al., ideR, An essential gene in mycobacterium tuberculosis: role of

IdeR in iron-dependent gene expression, iron metabolism, and oxidative stress response.

Infection and Immunity, 2002. 70(7): p. 3371-81.

164. Pohl, E., R.K. Holmes, and W.G. Hol, Crystal structure of a cobalt-activated diphtheria

toxin repressor-DNA complex reveals a metal-binding SH3-like domain. Journal of

Molecular Biology, 1999. 292(3): p. 653-67.

165. Que, Q. and J.D. Helmann, Manganese homeostasis in Bacillus subtilis is regulated by

MntR, a bifunctional regulator related to the diphtheria toxin repressor family of proteins.

Molecular Microbiology, 2000. 35(6): p. 1454-68.

156

166. Guedon, E., et al., The global transcriptional response of Bacillus subtilis to manganese

involves the MntR, Fur, TnrA and sigmaB regulons. Molecular Microbiology, 2003.

49(6): p. 1477-91.

167. Lieser, S.A., et al., DNA-binding and oligomerization studies of the manganese(II)

metalloregulatory protein MntR from Bacillus subtilis. Biochemistry, 2003. 42(43): p.

12634-42.

168. Golynskiy, M.V., et al., Metal-induced structural organization and stabilization of the

metalloregulatory protein MntR. Biochemistry, 2005. 44(9): p. 3380-9.

169. Stoll, K.E., et al., Characterization and structure of the manganese-responsive

transcriptional regulator ScaR. Biochemistry, 2009. 48(43): p. 10308-20.

170. Spiering, M.M., et al., Metal stoichiometry and functional studies of the diphtheria toxin

repressor. Proc Natl Acad Sci U S A, 2003. 100(7): p. 3808-13.

171. Pohl, E., et al., Structures of three diphtheria toxin repressor (DtxR) variants with

decreased repressor activity. Acta Crystallogr D Biol Crystallogr, 2001. 57(Pt 5): p. 619-

27.

172. D'Aquino, J.A., et al., Mechanism of metal ion activation of the diphtheria toxin

repressor DtxR. Proc Natl Acad Sci U S A, 2005. 102(51): p. 18408-13.

173. Chou, C.J., et al., Functional studies of the Mycobacterium tuberculosis iron-dependent

regulator. Journal of Biological Chemistry, 2004. 279(51): p. 53554-61.

174. Wisedchaisri, G., R.K. Holmes, and W.G. Hol, Crystal structure of an IdeR-DNA

complex reveals a conformational change in activated IdeR for base-specific interactions.

Journal of Molecular Biology, 2004. 342(4): p. 1155-69.

157

175. Golynskiy, M.V., et al., Metal binding studies and EPR spectroscopy of the manganese

transport regulator MntR. Biochemistry, 2006. 45(51): p. 15359-72.

176. Glasfeld, A., et al., Structure of the manganese-bound manganese transport regulator of

Bacillus subtilis. Nature Structural Biology, 2003. 10(8): p. 652-7.

177. DS, P., slation initiation at non-AUG triplets in mammalian cells. J Biol Chem, 1989.

264(9): p. 5.

178. Blattner FR, P.G.r., Bloch CA, Perna NT, Burland V, Riley M, Collado Vides J, Glasner

JD, Rode CK, Mayhew GF, Gregor J, Davis NW, Kirkpatrick HA, Goeden MA, Rose DJ,

Mau B, Shao Y., The complete genome sequence of Escherichia coli K-12. Science, 1997.

277(5331): p. 1453-62.

179. Kliegman, J.I., et al., Structural basis for the metal-selective activation of the manganese

transport regulator of Bacillus subtilis. Biochemistry, 2006. 45(11): p. 3493-3505.

180. Golynskiy, M.V., et al., Metal-induced structural organization and stabilization of the

metalloregulatory protein MntR. Biochemistry, 2005. 44(9): p. 3380-3389.

181. Pennella, M.A. and D.P. Giedroc, Structural determinants of metal selectivity in

prokaryotic metal-responsive transcriptional regulators. Biometals, 2005. 18(4): p. 413-

428.

182. Clarkson, S.M., et al., The Elemental Sulfur-Responsive Protein (SipA) from the

Hyperthermophilic Archaeon Pyrococcus furiosus Is Regulated by Sulfide in an Iron-

Dependent Manner. Journal of Bacteriology, 2010. 192(21): p. 5841-5843.

183. Trubetskoy, D.O., et al., Purification of proteins specifically binding human endogenous

retrovirus K long terminal repeat by affinity elution chromatography. Journal of

Chromatography A, 2002. 976(1-2): p. 95-101.

158

184. Moll, J.R., et al., Magnesium is required for specific DNA binding of the CREB B-ZIP

domain. Nucleic Acids Research, 2002. 30(5): p. 1240-6.

185. Bailey, T.L., et al., MEME SUITE: tools for motif discovery and searching. Nucleic

Acids Research, 2009. 37: p. W202-W208.

186. Crooks, G.E., et al., WebLogo: A sequence logo generator. Genome Research, 2004.

14(6): p. 1188-1190.

187. Bender, K.S., et al., Analysis of a ferric uptake regulator (Fur) mutant of Desulfovibrio

vulgaris Hildenborough. Appl Environ Microbiol, 2007. 73(17): p. 5389-400.

188. Rodionov, D.A., et al., Reconstruction of regulatory and metabolic pathways in metal-

reducing delta-proteobacteria. Genome Biol, 2004. 5(11): p. R90.

189. Robb, F.T., et al., Genomic sequence of hyperthermophile, Pyrococcus furiosus:

implications for physiology and enzymology. Methods Enzymol, 2001. 330: p. 134-57.

190. Schut, G.J., et al., Whole-genome DNA microarray analysis of a hyperthermophile and

an archaeon: Pyrococcus furiosus grown on carbohydrates or peptides. Journal of

Bacteriology, 2003. 185(13): p. 3935-47.

191. Kawakami, R., et al., Oxidative stress response in an anaerobic hyperthermophilic

archaeon: presence of a functional peroxiredoxin in Pyrococcus horikoshii. J Biochem,

2004. 136(4): p. 541-7.

192. Wardman, P. and L.P. Candeias, Fenton chemistry: an introduction. Radiat Res, 1996.

145(5): p. 523-31.

193. Pohl, E., R.K. Holmes, and W.G. Hol, Crystal structure of the iron-dependent regulator

(IdeR) from Mycobacterium tuberculosis shows both metal binding sites fully occupied.

Journal of Molecular Biology, 1999. 285(3): p. 1145-56.

159

194. Kosman, D.J., Molecular mechanisms of iron uptake in fungi. Mol. Microbiol. , 2003. 47:

p. 1185–1197.

195. Jongbloed, J.D., et al., Two minimal Tat translocases in Bacillus. Molecular

Microbiology, 2004. 54(5): p. 1319-25.

196. Baichoo, N., Wang, T., Ye, R., and Helmann, J.D., Global analysis of the Bacillus

subtilis Fur regulon and the iron starvation stimulon. Mol Microbiol, 2002. 45: p. 1613-

1629.

197. Grifantini, R., Sebastian, S., Frigimelica, E., Draghi, M., Bartolini, E., Muzzi, A., et al.,

Identification of iron-activated and -repressed Fur-dependent genes by transcriptome

analysis of Neisseria meningitidis group B. Proc Natl Acad Sci U S A, 2003. 100: p.

9542-47.

198. Baichoo, N., et al., Global analysis of the Bacillus subtilis Fur regulon and the iron

starvation stimulon. Molecular Microbiology, 2002. 45(6): p. 1613-29.

199. Louvel, H., I. Saint Girons, and M. Picardeau, Isolation and characterization of FecA-

and FeoB-mediated iron acquisition systems of the spirochete Leptospira biflexa by

random insertional mutagenesis. J Bacteriol, 2005. 187(9): p. 3249-54.

200. da Silva Neto, J.F., et al., Fur controls iron homeostasis and oxidative stress defense in

the oligotrophic alpha-proteobacterium Caulobacter crescentus. Nucleic Acids Research,

2009. 37(14): p. 4812-25.

201. Holmes, K., et al., Campylobacter jejuni gene expression in response to iron limitation

and the role of Fur. Microbiology, 2005. 151(Pt 1): p. 243-57.

202. Parker, D., et al., Identification of a Dichelobacter nodosus ferric uptake regulator and

determination of its regulatory targets. J Bacteriol, 2005. 187(1): p. 366-75.

160

203. Honsa, E.S. and A.W. Maresso, Mechanisms of iron import in anthrax. Biometals, 2011.

24(3): p. 533-545.

204. Jittawuttipoka, T., et al., Mutations of ferric uptake regulator (fur) impair iron

homeostasis, growth, oxidative stress survival, and virulence of Xanthomonas campestris

pv. campestris. Archives of Microbiology, 2010. 192(5): p. 331-9.

205. Mey, A.R., et al., Iron and fur regulation in Vibrio cholerae and the role of fur in

virulence. Infection and Immunity, 2005. 73(12): p. 8167-78.

206. Gilbreath, J.J., et al., Change Is Good: Variations in Common Biological Mechanisms in

the Epsilonproteobacterial Genera Campylobacter and Helicobacter. Microbiology and

Molecular Biology Reviews, 2011. 75(1): p. 84-+.

207. Touati, D., Iron and oxidative stress in bacteria. Archives of Biochemistry and

Biophysics, 2000. 373(1): p. 1-6.

161

APPENDIX

Table A1. Table of abbreviations

Abbreviation Definition Methods SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis EMSA Electrophoretic mobility shift assay qPCR Quantitative Polymerase Chain Reaction ICP-MS Inductively coupled plasma-mass spectrometry Chemicals IPTG Isopropyl -β-D-thiogalactopyranoside DTT Dithiothreitol BPS Bathophenanthroline disulfonate Protein names Fur Ferric uptake regulator DtxR Diphtheria toxin repressor Ftr1 Iron permease FeoAB Ferrous transporter Gene names fur Ferric uptake regulator, PF1194 dtxR Diphtheria toxin repressor, PF0851 ftr1 Iron permease, PF0723 feoAB Ferrous transporter, PF0857-PF0858 sipA Sulfur-induced protein A, PF2025 sufCBD FeS assembly proteins, PF1287-PF1285 aor Aldehyde:ferredoxin oxidoreductase, PF0346 pyrF Orotidine 5'-phosphate decarboxylase, PF1114 Strain names COM1 ΔpyrF COM1C2 ΔpyrF::pyrF FUR ΔpyrF Δfur DTXR ΔpyrF ΔdtxR::pyrF FTR1 ΔpyrF Δftr1 Miscellaneous terms LB Luria -Bertani broth YEM Yeast extract, maltose-based medium O.D Optical density ORF Open reading frame UOR Upstream of ORF region

162